Cataloguing Culture: Legacies of Colonialism in Museum Documentation 9780774863926, 9780774863940, 9780774863957, 9780774863964

Citation preview

Cataloguing Culture

Hannah Turner

Cataloguing Culture Legacies of Colonialism in Museum Documentation

© UBC Press 2020 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without prior written permission of the publisher. Library and Archives Canada Cataloguing in Publication Title: Cataloguing culture : legacies of colonialism in museum documentation / Hannah Turner. Names: Turner, Hannah, author. Description: Includes bibliographical references and index. Identifiers: Canadiana (print) 20200211021 | Canadiana (ebook) 20200211242 | ISBN 9780774863926 (hardcover) | ISBN 9780774863940 (PDF) | ISBN 9780774863957 (EPUB) | ISBN 9780774863964 (Kindle) Subjects: LCSH: National Museum of Natural History (U.S.) – Case studies. | LCSH: Museums – Collection management – Case studies. | LCSH: Ethnological museums and collections – Case studies. | LCSH: Indians of North America – Material culture – Case studies. | LCSH: Museums and Indians – Case studies. Classification: LCC GN406 .T87 2020 | DDC 069/.4 – dc23

UBC Press gratefully acknowledges the financial support for our publishing program of the Government of Canada (through the Canada Book Fund), the Canada Council for the Arts, and the British Columbia Arts Council. This book has been published with the help of a grant from the Canadian Federation for the Humanities and Social Sciences, through the Awards to Scholarly Publications Program, using funds provided by the Social Sciences and Humanities Research Council of Canada. Set in Myriad and Garamond by Artegraphica Design Co. Ltd. Copy editor: Barbara Tessman Proofreader: Helen Godolphin Indexer: Marnie Lamb Cover designer: Martyn Schmoll UBC Press The University of British Columbia 2029 West Mall Vancouver, BC V6T 1Z2 www.ubcpress.ca

To my family

A Po rt r a i t o f a D r awe r For whatever reason, wooden furniture has at various points in time been a metaphor for knowledge – as if there are large filing cabinets that sit in our minds, pregnant with intellectual possibilities. Physical furniture, like card catalogues and drawers, filled with information about books, objects, or even one’s own research notes, have likewise been seen as tools for producing new knowledges. Even museums themselves were occasionally called “Cabinets of Curiosity.” For this reason, the image here is both peculiar and dramatic. The drawer, placed carefully on a stool, is seated as if it is looking straight into the camera. It was taken some time in 1890 in the United States National Museum (now, the National Museum of Natural History) at the Smithsonian Institution in Washington, DC. It was created in silver gelatine and also cyanotype; the former image is reprinted here. There are other images like it – drawers, chests, wooden storage cabinets. Perhaps they were chosen as ideal subjects for the camera because of their stillness, or perhaps because they were marvels of bureaucratic modernity, or perhaps they were for sale. The drawer may have held objects, but more likely it contained documents and registers of collections information or other routine work. Whatever the case, it is an unusual but effective visualization for the goal of this book – to centre the technologies and practices that make up knowledge work in museums. I encourage readers to imagine why the material practices of documentation continue to be important for all of the communities and individuals whose objects and information still reside upon documents and within museum cabinetry.

USNM storage drawer | Smithsonian Institution Archives, Image MNH-4477

Contents

List of Figures and Tables / x Acknowledgments / xii List of Abbreviations / xiv Introduction: “The Making of Specimens Eloquent” / 3

1 Writing Desiderata: Defining Evidence in the Field / 28



2 On the Margins: Paper Systems of Classification / 65



3 Ordering Devices and Indian Files: Cataloguing Ethnographic Specimens / 87



4 Pragmatic Classification: The Routine Work of Description after 1950 / 125



5 Object, Specimen, Data: Computerization and the Legacy of Dirty Data / 157 Conclusion: A Museum Data Legacy for the Future / 184 Notes / 194 Bibliography / 216 Index / 228

Figures and Tables

Figures

1 Plans for the US National Museum, 1878 / 32



2 Workroom for Indian ethnology, ca. 1890 / 67



3 First pages of anthropology ledgers, ca. 1859 / 72



4 Illustrations in anthropology ledgers, ca. 1884 / 76



5 Indian file / 88



6 Storage area in the US National Museum / 91



7 Walter Hough, 1908 / 108



8 Walter Hough and staff examine textiles from a new ethnological collection / 108



9 George D. McCoy in the Catalogue and Recording Office / 109



10 Frank M. Setzler prepares archaeological collections for study / 113



11 Typewritten catalogue card / 118



12 Catalogue card, 1965 / 119



13 History of Collection card / 121



14 Key to Collection card / 121



15 Computer room in the Automatic Data Processing Center, 1978 / 139



16 Tom McIntyre catalogues squirrel specimens / 139



17 Fred Collier and Jann Thompson looking at a SELGEM printout / 140



18 Inventory Worksheet / 145



19 Preferred index specimen names / 153

20 Catalogue card marked for terms for computerization / 165

Tables 1 List of observances and object types desired, from Holmes and Mason, Instructions to Collectors / 56

2 Ledger headings through time / 74

Figures and Tables

xi

Acknowledgments

This book has followed me from Toronto to Vancouver to the United Kingdom and back home again, and it carries the legacies of the work of many others who helped enormously. Writing it has taught me the importance of upsetting the authority that is so readily given to institutions, of thinking critically about the work and technologies used, and of making space for other ways of imagining. It has also taught me to challenge my own assumptions as a settler scholar – and that is not work I will ever be finished with. Thank you to everyone who helped or inspired me with their support, writing, and labour. At the Smithsonian and the National Museum of Natural History, I thank Candace Greene, who conducted the early research on museum catalogue fields and held a long career at the NMNH. Without her research and encouragement, this project would not have happened. Gwyneira Isaac, who was my research supervisor at the Smithsonian, helped make a large institution feel small and welcoming. Thanks are also due to Eric Hollinger, Carrie Beauchamp, and Felicia Pickering for their willingness to talk, their support, and for reviewing and adding comments on the final manuscript. Thank you also to the archivists at the National Anthropological Archive and the Smithsonian Institution Archive for initial research help and for tracking down images. I thank my colleagues and fellow Smithsonian researchers Diana Marsh, Adrian Van Allen, and Catherine Nichols for their own scholarship and friendship. At the University of Toronto in the Faculty of Information, Cara Krmpotich, as my doctoral supervisor, guided the

formation of these ideas and supported me in invaluable ways. Seamus Ross and Costis Dallas, my committee members, also gave important feedback and encouragement; and Quinn Dupont and Ashley Scarlett have been sources of support since our time as doctoral students. As a SSHRC Postdoctoral Fellow in the Making Culture Lab at the School of Interactive Art and Technology at Simon Fraser University, I spent much of my time preparing my dissertation to become this book and found many friends and colleagues. Kate Hennessy has been a great mentor and collaborator, and her work continues to inspire me. I also spent my time editing the first draft of this book at the University of Leicester in the School of Museum Studies as a lecturer, and my colleagues there contributed time and dedication to the project: thank you to Sandra Dudley, Simon Knell, and Ross Parry for reading earlier drafts; and to Nuala Morse and Laura Gibson, who gave me comments, tea, wine, and friendship throughout my time in the UK. At UBC Press, I am thankful to Darcy Cullen and Ann Macklem, who have ushered me through my first monograph publication with care. I am especially grateful for the skilful editing of Barbara Tessman, who has, magically, made my writing sound coherent, and who also managed to keep my voice intact throughout. Thank you to my anonymous reviewers, who took the time to give me important and constructive feedback. Of course, as always, all mistakes are my own. Most importantly, thank you to my family and friends. In ways both mundane and spectacular, they have made all the difference. The original funding for this research was provided by the Social Sci­ ences and Humanities Research Council of Canada (SSHRC) and fur­ther funds were provided by the University of British Columbia Scholarly Pub­ lication Fund. Parts of this book have been the foundations for three articles, published in Museum Anthropology (2016), Knowledge Organization (2017), and Cataloging and Classification Quarterly (2015).

A c k n ow l e d g m e n t s

xiii

Abbreviations

ADP

Automatic Data Processing

BAE

Bureau of American Ethnology

CMS

content management system

HNAI

Handbook of North American Indians series

HRAF

Human Relation Area Files

MSC

Museum Support Center

NAGPRA

Native American Graves Protection and Repatriation Act

NMAIA

National Museum of the American Indian Act

NMNH

National Museum of Natural History

NPS

National Parks Services

OCIO

Office of the Chief Information Officer

SIMA

Summer School in Museum Anthropology

SOA

Smithsonian Office of Anthropology

USNM

United States National Museum

Cataloguing Culture

Introduction

“The Making of Specimens Eloquent”

The making of specimens eloquent demands more than the mere objects. Besides collections, cases, cabinets, alcoves, exhibition halls, study series, dust proof drawers and constant watch care, there must be illustrations, drawings, photos, paintings, portraits, portfolios, and picture galleries. There must also be information, card catalogues, labels, reference lists, pamphlets, library, and other sources of knowledge. The public lecture, uniting the specimens, the pictures, and the label with the eloquence of the lecturer and the sympathy of many hearers, completes the story. Everyone who has the interests of the science of our species at heart sufficiently to save the material witness, will confer the greatest benefit by being sure of the data accompanying each specimen. – Otis Mason, curator of ethnology, United States National Museum, 1906 Each Federal agency and each museum which has possession or control over holdings or collections of Native American human remains and associated funerary objects shall compile an inventory of such items and, to the extent possible based on information possessed by such museum or Federal agency, identify the geographical and cultural affiliation of such item ... The term “documentation” means a summary of existing museum or Federal agency records, including inventories or catalogues, relevant studies, or other pertinent data for the limited purpose of determining the geographical origin, cultural affiliation, and basic facts surrounding acquisition and accession of Native American human remains and associated funerary objects subject to this section. – Excerpt from the Native American Graves Protection and Repatriation Act, 1990

What are the legacies of colonialism? How do colonial relations present themselves through time and in practice? This book does not present an easy answer to these questions, but it points to the material, documentary practices of ethnographic museum work as a key site in the production of continued colonial legacies. For example, alongside American legislation aimed at improving originating communities’ access to their cultural heritage and facilitating repatriation requests, we find that our procedures and policies for doing so are often moored in past ideologies or epistemic loyalties. What are the material and historical practices that continue to affect current ethical considerations? Museum records are often taken to be neutral or privileged sources of knowledge, but they are both contextual and historical, as are many bureaucratic practices. As many practitioners, experts, and scholars know, to return or de-accession objects, one must investigate the documentation that was collected with the object or that was created by the institution. Repatriation work is a complex process of relationship building, advocacy, activism, fundraising, and more. It also requires a lot of “paper” work, digging into archives and museum catalogues to establish claims of ownership and “authenticity.” As I will show in this book, institutional knowledge, particularly in museums, exists in the work of record keeping, data collection, and (today) digitization. This work includes naming, standardizing, classifying, and excluding, and these practices have legacies that reach far beyond the present. In this book, I suggest that the history of anthropology is also a history of paper media technologies that have cocreated our understanding of the past. This study is a deep investigation into the pressing issues that arise when we address the work of documentation in museums back through time. To do this, I take as the major case study of this book the Smithsonian’s National Museum of Natural History (NMNH),1 which is where I conducted my doctoral work.2 There, in the Department of Anthropology, I found a history of museum bureaucracy in ledger books, card catalogues, staff members’ memories and files, and the institutional archives. As I will argue, the museum discursively constructed ethnographic specimens as it recorded and documented them. This book is a history of classification and documentation in one ethnographic museum, but it pays attention to the constitution of material-culture-as-data to historicize the collection of data about human beings more broadly. What made a “good” specimen was markedly different in 1880 than it was in 1908, or 1970. In nineteenth-century North America, objects and their associated information were brought in,

4

Introduction

recorded in ledger books, and catalogued in large indexes as data; and these documentation practices came to form fundamental concepts in anthropology and material culture studies. I make no claims that this process was or is an all-encompassing system of description. Instead, I am most intrigued when classifications cease to work and when individuals themselves encounter and resist what they see as bureaucratic inefficiencies. To study these moments requires a kind of “studying up” that turns the normalized and stable into something that is to be investigated and opened.3 This approach is as much a personal philosophy as it is an academic task, one that has driven my work and brought me to study these historical cabinets and “old” media technologies in museums as prolific producers of categories about human beings that have present political and ethical importance. When museum documentation is seen as neutral – in legal repatriation proceedings, for example, or in the US Native American Graves Protec­tion and Repatriation Act (NAGPRA) excerpt above – we must question what the origins of this objectivity are, where it comes from, and how it stands to last. Throughout this study, I consider how intellectual categories and ideas were established and became normalized in the study of material culture. Documentation media have not been subjected to the same criticism directed at visual media like photography or film. Yet, the collecting list, the paper register, and the card catalogue were foundational media technologies in the development of the anthropological discipline in the midnineteenth and twentieth centuries, and they created the conditions of possibility for contemporary digital databases. The ways that ethnologists (and later, anthropologists) made use of paper and digital technologies worked to reinforce the authority and integrity of scientific colonialism. I argue that we must be attentive to the historical and grounded practices that have organized material culture as “specimens,” while at the same time remaining cognizant of the socio-political ramifications when conducting histories such as these. To understand how colonialism operates as both a productive and reductive force, it is necessary to investigate how categories were applied to material culture and became routinized through bureaucratic documentation in collecting institutions. Early North American museum ethnologists focused on the preser­ vation of Indigenous cultures from all over the world by collecting their belongings. At the same time, as many examples in this book will show, nar­ratives belonging to these Indigenous communities were often excluded and considered to be unnecessary for the scientific pursuit of the study of

“The Making of Specimens Eloquent”

5

humankind.4 In North America in particular, this was, in part, intentional: a way to claim resources and lands for colonial states. It was also unintentional or, rather, so embedded into practice that it rarely was seen for what it was. The origins of much documentation practice in museums used ancestors, belongings, and collections as the “data” of human history – and I believe this is important to consider. Histories such as these are present in almost every ethnographic museum in the world, and many other scholars and writers have covered the implications comprehensively.5 For example, Marisa Elena Duarte and Miranda Belarde-Lewis articulate an ethical stance about the need for understanding how colonialism is continually practised in information and heritage institutions: For at least half a century, catalogers have struggled with how to catalog and classify Native American and Indigenous peoples’ materials in library, archive, and museum collections. Understanding how colonialism works can help those in the field of knowledge organization appreciate the power dynamics embedded in the marginalization of Native American and In­digenous peoples’ materials through standardization, misnaming, and other practices. The decolonizing methodology of imagining provides one way that knowledge organization practitioners and theorists can acknowledge and discern the possibilities of Indigenous community-based approaches to the development of alternative information structures.6

For Duarte and Belarde-Lewis, decolonization is a process by which In­ digenous and non-Indigenous scholars can imagine new realities, ontologies, practices. As they suggest, naming is one of the key ways in which colonialism and white superiority continue to be oppressive for Native American and Indigenous peoples. Other ways that the subjugation of Native knowledge systems is practised are in the delinking of knowledges from greater ontological systems and in the persistent idea that, if we change any of the ways we classify and standardize something, we are uprooting some kind of moral order handed down from on high.7 We need to recognize that, “in the everyday sense, the power to name is a way of organizing, of itemizing, of making information and knowledge accessible to both a specific and imaginary constituent audience.”8 As a white settler scholar, rather than attempting to reproduce Indigenous knowledges, histories, or stories, I turn my attention to the information practices that enabled particular epistemologies to be taken “as is” and remain

6

Introduction

durable in the history of anthropology and museums. In doing so, I do look at records and objects that may have been taken from communities with­ out their consent. I remain cognizant that there are other, important ways to tell stories regarding ancestors and belongings, and that for the longest part of the history of the disciplines, the theories and histories of anthropology were created in a top-down fashion by white male scholars (although there are important exceptions to this generalization).9 Yet, as Margaret Bruchac recounts skilfully in her book Savage Kin, early Native inform­ants were actively managing early ethnologists and anthropologists, and even mitigating potential harm.10 She argues for us to make the “Indigenous intellectual and social contributions to the foundational knowledges of anthropology far more visible.”11 What I work to show is how museum documentation, as an embedded institutional practice, helped cement narratives that have assisted in erasing, or at least hiding, those subaltern narratives brought to the fore in works like Bruchac’s. The subtitle of this book – Legacies of Colonialism in Museum Documen­ tation – connotes how these long-standing and pervasive ideologies are still at work today. One of my concerns is the idea of “legacy data,” a term used for information attached to museum objects in their records. This information includes names, places, and stories that collectors and museum workers assigned to objects in the past. The term is also a provocation to think crit­ ically about how documentation embeds certain narratives and occludes others. Data (big or small) are historical and are variously valued, and, as I have found, they can be both “dirty” and “clean.” Principally, the word “data” usually carries with it an assumption of veracity and reverence for the possibilities of impartial, omniscient technologies. Yet the ways in which we craft knowledge from “data” are not impartial, and it is incorrect to assume that data repositories are detached from their histories. As Daniel Rosenberg has argued, the theoretical underpinnings of “data” can be traced to the seventeenth century. The term was linked to ideas about argumentation and knowledge, as it was recognized that the accumulation of data could make the unknown known.12 “Data,” Rosenberg writes, “has no truth.”13 Such assumptions about data are based on similar concepts about the objectivity of facts and truth in the sciences. Yet data are constructed and are historical; they are not transparent or self-evident.14 We need to interrogate the histories of “raw” data across times, geographies, and circumstances, asking how different disciplines have imagined or crafted their objects and evidence.15

“The Making of Specimens Eloquent”

7

This book raises two key ideas: first, that the origins of media technologies used to catalogue humanity are an under-examined yet crucial history; and second, that contemporary claims to knowledge authority in cultural or memory institutions are built upon the standards that have organized the remains and material objects of North American Indigenous communities. Documents – what Annelise Riles has called “the artifacts of modern knowledge” – shaped the development of anthropology but also our relationship to understanding history and human beings.16 As material culture became data in the museum and was mined as a resource for scientific knowledge extraction, the legacy of past practice and epistemology was written into documentation as categories, naming conventions, incorrect tribal affiliations, and more. As communities reclaim their heritage, and museum staff accommodate this reclaiming, both encounter these documentary traces of imperialism that continue to frame the legal and ethical approaches to repatriation requests.17 What, then, is the history of museum data? The first epigraph at the beginning of this introduction was taken from the annual report of Otis Tufton Mason, who was the curator of ethnology at the Smithsonian’s United States National Museum (USNM; now the NMNH) from 1884 to 1908. This excerpt, among many others in his annual reports over his tenure, is about the systematic data-centred approach to the burgeoning field of anthropology. Legitimizing anthropology as a true science, Mason argued, required the objects of science. In a kind of agential-realist intellectual manoeuvre,18 it required crafting and manipulating a series of technologies and practices that have come not only to define the field but also to form a general approach to creating knowledge from what he saw as “material witnesses.” The second excerpt is from American repatriation legislation known as the Native American Graves Protection and Repatriation Act, passed into law in the United States in 1990. Although the discipline of anthropology has changed significantly since 1884, there are concerning similarities between the kinds of claims made in Otis Mason’s statement and those made in the NAGPRA. Both contain inherent assumptions: first, that the museum is the authoritative source of information about these objects and, in many cases, human remains; and, second, that this authority is produced by regulated and, to some extent, standardized systems.19 These similarities, despite the intervening century, stimulated my interest into the mundane and routine aspects of knowledge work in museums and anthropology, and how data inherit the ideologies and legacies of colonialism.

8

Introduction

In the mid-nineteenth century, naturalists and government employees were interested in collecting the material culture of Indigenous peoples in North America. Through a variety of methods, objects were bought, sold, and taken from communities that were actively engaging in complex cultural exchanges. The USNM was an official government collection of all kinds of items – biological, paleontological, geological, and ethnological. The goal of the museum was the creation of large “data” sets of all life, and it guided the collection of material culture from all over the world to aid in this endeavour. The Bureau of American Ethnology (BAE) was established as part of the USNM in 1879, and the Anthropology Division was officially created within the USNM in 1883. This began the bureaucratic and data-rich life of material culture collections, and the legacies of colonial naming conventions, terminologies, and methodologies devised at that time still impact how these collections are used today. The use of seemingly mundane recording technologies worked to reinforce the authority of the Smithsonian Institution as the source of information about objects and histories. Revealing how anthropology has created, classified, and catalogued its objects opens up these histories in a new way.

Anthropological Classification and Museum Objectivities In 1966, the social anthropologist Mary Douglas published the book that would make her well known – Purity and Danger. This was a detailed discussion on holiness, supported by her understanding of the Christian faith, her fieldwork, and other anthropological examples. As she says, “Dirt is matter out of place. It implies two conditions: a set of ordered relations and a contravention of that order. Dirt then, is never a unique, isolated event. Where there is dirt, there is a system. Dirt is the by-product of a systematic ordering and classification of matter insofar as ordering involves rejecting inappropriate elements.”20 In her book she compares “Western” ritual habits around holiness to those of smaller “primitive” cultural groups, as a way to denormalize the practices and beliefs of her contemporaries in the academy and “Western” society. She does this by comparing different systems of ritual food avoidance and taboo, but her book is more than an understanding of why Christians, Jews, and the Nuer avoid particular foods. She theorized a world full of classifications – one where anomalies, outsiders, monsters were necessary parts of the “system.” She showed how naming affects our entire understanding of things, how we place ourselves in hierarchies that

“The Making of Specimens Eloquent”

9

are relative, not absolute. “Holiness” implies cleanliness, and, beyond the old adage linking the two, Douglas argues that this connection is essential for understanding how we make sense of the world. Her thesis on dirt and purity was simple: that our ideas of what fits and what does not build our worlds. Douglas was working within a paradigm of colonial anthropology, something we must not forget. But her work laid the foundation for a serious and investigative study of categories as relative things, arguing that what counts as “discordant” and “taboo” is a complex aspect of how things are named communally. As she writes, “Ambiguous [things] tend to be treated as if they harmonized with the rest of the pattern. Discordant ones tend to be rejected. If they are accepted, the structure of assumptions has to be modified. As learning proceeds, objects are named. Their names then affect the way they are perceived next time: once labelled they are more speedily slotted into pigeon-holes in the future.”21 One of her most salient conclusions was that “Western” society is not at the top of a hierarchy of classification where everything has its own place. She argued instead that every society has complex, and not purely logical, characterizations of what “counts” as an object in a pattern. Her work was important because it directly questioned the long-held belief that permeated many disciplines at the time: Western assumptions of order were superior to all else. European science was a specific kind of science, characterized by an orientation toward finding global truths derived from natural historical descriptions.22 As the museum scholar Tony Bennett argues, classification represented a new condition of possibility and a new kind of evidence: that of difference.23 Classificatory tables, hierarchies, and arrangements based on visual characteristics are all indicative of what has been called a Classical episteme, where these modes of ordering presented “simplified, yet utterly verifiable knowledge.”24 Mary Louise Pratt has argued that the “natural history approach” to the study of human beings and their material culture was made possible by developments in the classification of natural history specimens that originated with the publication of Linnaeus’s “Systema Natura” in 1735. Indeed, as Pratt notes, Europe’s first major international scientific expedition was to determine (once and for all) the shape of the earth. The “descriptive apparatus of natural history,” as Pratt calls it, defined the approach to gathering information on this expedition, and future ones like it. This apparatus made it possible to document the encounters with the world that early European travellers faced, and to create a kind of order from chaos through classification.25

10

Introduction

Kevin Hetherington, echoing Michel Foucault, has argued that relationships of similitude were key ordering devices used by museums in the nineteenth century, and the use of the classificatory table presupposed a different “gaze,” which constructed an entirely different subject. Rather than being opposites, similitude and difference enabled early ethnologists both to group objects collected from the field that shared similar characteristics and to establish series of these objects based on their differences.26 These differences were often based on functional characteristics that were assessed visually. The process of linking objects within larger series of evolutionary progress is often called the typological tradition. Preceded by archaeological work that created grand typologies of artifacts to understand the temporal relationships of objects found in the ground, typological arrangements make visible an evolutionary scheme through an object-centred pedagogy. The “type specimen” in botany, for example, was considered by natural historians and scientists to be representative of an entire class or group of specimens, and was carefully constructed to emphasize the average traits of the object.27 This tradition was made explicitly visible in museum displays of anthropological collections. An early example of this is Pitt Rivers and his organizational attitude to collections in the museum in Oxford that bears his name. As part of what David Jenkins calls an “emerging ethos of scientific reality,” this evolutionary paradigm is diametrically opposed to earlier modes of organization in early wunderkammern, or “cabinets of curiosity,” whose displays emphasized the unique or monstrous objects collected during travels.28 Objects were no longer, to borrow Mary Douglas’s term, “deviant” monstrosities, but had transformed into suitable specimens and objects of evidence by being aligned in a legible series or a logical progression. Tony Bennett has argued that this shift allowed museum objects to perform a new function as observable phenomena to be studied by science, and to circulate between collections and among museums.29 Thinking of objects as evidence enabled a survey of the historical and social processes in the study of humankind’s evolution. The foundations and developments of natural history thus made it possible (and plausible) to examine the natural state of humanity itself. A new field of visibility was created, and the methods of observing the natural world provided the key to understanding its wonders and minimizing its chaos.30 Order, as Foucault writes, was a model of rationality.31 The use of the Linnaean system had clear and measurable effects on the study of people in the context of natural history, and, as Pratt notes, the

“The Making of Specimens Eloquent”

11

system “epitomized the continental, transnational aspirations of European science.”32 Indeed, early anthropological investigations formed hypotheses about the world in this context, with the Linnaean classification system acting as a rubric that allowed scholars to structure their findings, not just about plants, but about humankind as well. As Pratt notes, this approach was reflected in Linnaeus’s work: “Linnaeus posited among the quadrupeds a single category homo ... and drew a single distinction between homo sapiens and homo monstrous.”33 Within the context of Linnaean classification, any form that falls outside of the internal logic is considered a singularity.34 As the principle of evolution was popularized, however, the distinction between the monstrous and normal became a trivial one – the outliers, or “freaks,” became the “missing links.”35 Early naturalists therefore often presented objects as pieces of evidence that became key components in telling the story of science, progress, and “man.”36 Further, objects collected from colonies were important because of the primacy of visual instruction. Tony Bennett argues that “eye-knowledge arose from the distinctive epistemological concerns of the historical sciences in their claims to be able to decipher the meanings of objects and, thereby, to challenge the text-based narratives of biblical and humanistic scholarship.”37 The reliance on visual pedagogies, or “ocularcentrism,” and the quest for objectivity in the sciences, influenced anthropological approaches to understanding culture. This way, as Bennett argues, “the whole of the material world could be lined up and placed before the eyes in a manner which allowed each display to tell its own story, seemingly without the need for textual mediation.”38 This offered a way to represent the progress from savage to barbarian to “civilian,” enabling ethnological displays to validate the utopian ideals of nineteenth-century elites above all else.39 This image of utopia ultimately saw a future where the disappearance of Indigenous populations was imminent. Lorraine Daston and Peter Galison have sought to understand how visual modes of thought or metaphysical and epistemological realities come into being. They argue that there have been several modes of seeing in the sciences and that this variety has had an effect on how knowledge is created and how science is “done.”40 They trace the development of three modes of “objectivity”: truth-to-nature, mechanical objectivity, and trained judgment. Daston and Galison make the case that objectivity, or knowing the world through seeing, is a cultivated practice, one that has shaped the prac­ tice of scientific observation. They examine scientific atlases and catalogues

12

Introduction

as visual evidence, and sites of epistemics and “ethos,” where visual habits are “also expressions of epistemological loyalties.”41 The creation of a scientific image, they argue, is the creation of a scientific self. As objective observation thus became the dominant empiricism of the sciences, the articulation of the set of practices that involved the collection of type specimens in the sciences “reconciled scientific memory and amnesia.”42 Above all, objects could be made to stand in for, represent, or replace the human objects of inquiry, either visually or physically, and, thus, museum displays made ideal educational tools. This use of objects resulted in a strategic approach to collecting material culture, particularly in North Amer­ ica. In the United States from 1860 until 1900 – what has been called the Museum Period43 – a pervasive approach to understanding “other” cultures developed and was institutionalized through museum collecting and cataloguing practices. The theoretical positions of early ethnologists were strongly empirical and retained the natural historical approaches of training within the biological and geological sciences. The first curators and cataloguers in museums were trained as naturalists but operated as amateur ethnologists.44 For example, Frederick Ward Putnam, in his association with the Harvard Peabody Museum, was trained within a particular kind of evidentiary regime, bringing to early anthropology a “respect for data as evidenced by authenticated collections.”45 The “data” of ethnological research were situated in the “field” and manifested as documents (field notes, observations) and museum collections. Fifty years of scholarship have considered the history of anthropology as a history of colonialism. James Clifford and George Marcus proposed a language of postcolonial and reflexive critique from which the discipline of anthropology has emerged. Serious engagements with the construction of culture through the academic inscription and “writing” of it have since abounded.46 Similarly, the view of museums as non-neutral knowledge institutions has received critical scholarly attention since at least the 1970s, and decades of postcolonial research and activism on the part of Indigen­ ous communities have reconstructed the museum as a site of harm, contestation, contact, cultural negotiation, and potential healing.47 The history of anthropology in North America is also closely tied to the development of collections in museums,48 and the intellectual categories crafted by colonialism and used to exclude communities and people have been criticized for some time. Yet little work has addressed how the epistemic loyalties of colonial collecting practices became embedded into the everyday practices

“The Making of Specimens Eloquent”

13

of institutions. As Geoffrey Bowker and Susan Leigh Star have done,49 I bring attention to the ways in which classification can be used to discriminate, and I plot the historical underpinnings of systems of categorization. It is my goal to show that ideas about evidence are historically situated and arose out of the communities and techniques of scientific-ethnological practice in the long nineteenth century.50 Many of these ideas are influenced and exemplified by recent work in media history and information studies.51 These fields have broadly called us to pay attention to the infrastructures of knowledge production that influence a collective understanding of data-as-knowledge.52 Alex Wright’s and Markus Krajewski’s recent works on organizational devices and card catalogues constitute important histories of technology and information.53 Where Wright and Krajewski plot the origins of card catalogues in the library sciences, this book bridges these media histories with the history of anthropology and the development of paper organizational systems. Critiques of the history of colonialism and media have also focused on the representation of peoples and cultures through vision – namely, through display and photography.54 Brian Hochman’s recent work on the history of anthropological recording technologies, such as the phonograph, evaluates the history of visual and audio recording technologies alongside the history of early ethnology. It stands as an example of how media technologies and anthropology have influenced one another.55 Similarly, Rebecca Lemov examines a “lost” archive of punch cards at the Library of Congress. She found that these “little tools” of knowledge were used by twentieth-century cognitive anthropologists to record the dreams of Indigenous interviewees.56 Here, I look more closely at the kinds of documents, ledgers, forms, and lists used to record material culture during a similar time period. Lastly, research and writing in postcolonial, feminist science studies have brought to media stud­ ies and history a grounded politics that seeks to upset what we see as normalized and routine. Scholars have also productively raised the notion of seeing museums and the study of material culture as a key development in the history of Western and colonial anthropology, and I build upon their critique. As I argue throughout this book, how we make knowledge relies on a complex system of individuals, technologies, and technical skill – all of which are compounded in the way documents are created and circulated. As an information studies and museum studies scholar, I am interested in describing the legacy of how colonial thought can become stabilized in material technologies and practices – what has been called “material durability.”

14

Introduction

Theorized much earlier by sociologists of science such as John Law, Michel Callon, and Bruno Latour, the concept of material durability has encouraged me to look into the small iterative ways that forms and epistemologies can remain stable despite their ability to disappear into the woodwork – or, if you like, cabinetry. “Durability” typically connotes objects that have material force, things that can be touched and pushed against. Relations can be made “durable” when they become embedded in “inanimate materials” such as buildings or texts.57 Put another way, social arrangements maintain their network relationship longer when they are made outside of the relation itself, when they are made physical.58 The concept of discursive stability is also useful in understanding the way relations are made “durable.” Discourses can come to have lasting effect in both physical environments and within practices: the work of individuals and their relations, that “hold together” in, for example, an organization. Through a Foucauldian lens, Law argues that certain ways of ordering are made also possible by certain discourses – the discourse sets the limits and the conditions of what is possible.59 Yet this reading of durable materials and discourses is complex. As Law explains, the walls of a prison are only as durable as the guards who watch them and the bureaucracies that order the relations between the guards and the prisoners, and indeed the relations between durable materials change depending on their location within a network, or their adoption into a new one.60 In her recent work, Ann Stoler has taken the concept of durability or duress as a capacity of colonial or imperial ideations to endure, to hold out and last, despite shape-shifting in form.61 For Stoler, imperial forms maintain a kind of legacy – which she calls “presence” – where the logics of imperialism continue to frame contemporary encounters. Looking at colonialism and temporality, as Stoler suggests, calls us to question how an ideology (or set of embedded practices) can remain materially durable. Effectively, I am interested in the small details and specifics of how colonial legacies present themselves in documentation technologies. The idea that cataloguing is as much a practice as it is a set of rules is not new; given this idea, we can see cataloguing as a performative knowledge process.62 As Annemarie Mol teaches us, such a conceptual starting point supposes that multiple realities are practised, or “done,” and that they are situated not in the realm of philosophy but in everyday, routinized work.63 Mol’s research was fundamental in examining the daily work of nurse practitioners to understand how diagnosis is negotiated between patients and medical professionals. As she argues, the action of this work is a kind of

“The Making of Specimens Eloquent”

15

performance enactment or performativity that is heterogeneous – that different lived realities are practised differently.64 As Clare Waterton has recently argued of archives, performance is something “done,” whereby “objects and categories are only really present in ‘the doing of them,’ [and] they have to be continually performed to exist at all.”65 The doing of the work, the creation of routine, and the focus in this book on the practices of cataloguing are therefore all relevant.66 Put simply, performativity is the way in which things change and are enacted in practice. These performances become naturalized, and the distinctions, politics, and genders enacted within often hide the political nature of actions.67 I build on this critical scholarship to consider the formulation of ethnological categories and classifications from a perspective that seeks to destabilize existing normative claims to evidence and objectivity. Throughout this book, I use the term “material durability” to refer to a feature of what can be called the “information infrastructure” of the anthropology catalogue at the NMNH. I argue that, despite decades of postcolonial research and revision, object names and classification terms seem to stick to existing object records, which were situated in the natural historical sciences of the nineteenth century. I also suggest that structured museum bureaucracies arose from the routine socio-technological practices of the everyday work of organizing, classifying, and cataloguing; and this process can impact In­ digenous peoples’ abilities to access and to ensure the return of their cultural heritage today. Through this study, I hope to raise essential questions for anthropologists, museum historians, and media scholars: How have we constructed the “data” of material culture? Also, for whom does this process matter, and why?

Outline of Chapters Understanding the collecting of objects and anthropological data in the nineteenth and early twentieth centuries depends on understanding the practice of collecting more broadly. In the context of this book, it also means recognizing that the Smithsonian and its museums were, and are, an arm of the American federal government. The Smithsonian was established in 1846. At the time, the practice of science, particularly natural history collecting, was well established. However, in the eighteenth and early nineteenth century, there was little in the way of standards of practice when it came to collecting objects – indeed, they often arrived at an institution haphazardly

16

Introduction

or from pre-existing personal collections. At mid-century, a number of collecting expeditions and world’s fairs resulted in a mass of objects from military men and budding ethnologists that came into museums, including the Smithsonian, in piles. Objects would enter a collection in boxes, often with little or no contextual information. At the Smithsonian, they were then recorded in large ledger books, which contained fields to guide the recording of ethnological information, but these fields were directly copied from those in other departments – for example, minerals, zoological objects, and ethnological artifacts and human remains were catalogued using the same documentation fields as these other departments to describe each object. When records were computerized at the Smithsonian in the 1960s, they made use of the same terms and ideologies from a century earlier, while data took on new meanings and values. The chapters that follow take these media technologies and paper documentation tools as their objects of analysis. They are organized around the technologies that have framed the study of material culture: the field recording list, the ledger book, the card catalogue, the computerized inventory, and the database. Chapter 1, “Writing Desiderata: Defining Evidence in the Field,” demonstrates how, in the middle of the nineteenth century, the Smithsonian Institution worked to establish the very objectivities and categories by which we measure “human-made” things. Establishing normative claims to evidence was done through the collection of “field data,” a term transported from other sciences to ethnography. “Desiderata” were lists published in Smithsonian scientific correspondence – field guides and circulars – sent throughout North America and the world. These documents and lists specified the kinds of things that were seen to be “useful” to the burgeoning scientific practice of ethnology.68 This chapter plots the development of the “science” of ethnology from the mid-nineteenth to the early twentieth centuries, arguing that paper documents and circular lists were foundational technologies that helped craft contemporary objectivities around the study of the past. I show how the epistemic loyalties have changed through time, depending on what was desired, and when. I bring attention to how objects and information about people became evidence in the language of science at the time through the use of documentary tools. Staff made collections useful by connecting the documentation to the object through good, bureaucratic record-keeping systems. As a result of invitations for the collection of objects published in circulars, the Smithsonian received at least hundreds of specimens. The collecting guides detailed in this chapter are

“The Making of Specimens Eloquent”

17

important examples of how non-specialists and scientists envisioned what a best practice of museum collecting might be at the time. The guides made it possible for collectors who were non-experts in the field to work on behalf of the Smithsonian, acquire the desired objects, and preserve them at a scientific standard that the museum saw fit. As is evidenced in the rhetoric of these early circulars, without basic documentation, objects were of little or no research value. Objects that were collected in the field were documented and catalogued; through this process, they became scientifically meaningful and formed the basis of future ethnological research. Further, the circulars mirrored or explicitly used methodologies for collecting natural history specimens in the field. As a result, objects were crafted as “wellauthenticated,” scientifically sound specimens. The collecting guides give voice to the epistemic concerns of natural history, and the classificatory and ocularcentric gazes. The “descriptive apparatus” of natural history foregrounds the Smithsonian’s approach to collecting in the mid-nineteenth century, as evidenced by the kinds of description and data collection proposed in the field guides. The second chapter, “On the Margins: Paper Systems of Classifica­tion,” examines the ledger books used to record objects once they were in the museum. I consider these as paper technologies and highly idealized media forms; they were ideal “blanks” that established the institution’s authority, and staff recorded data in them after the objects were brought in and unpacked.69 The practice of writing objects down in these registration lists allowed them to be legible within the institution and across departments as specimens. These ledgers are not free from human error, yet, as this chapter shows, current museum work, including repatriation claims, often relies on the information recorded in them. In this chapter, I look to the specific collection and catalogue made by Timothy Dixon Bolles, the collector of what is known as the Hoonah Repatriation Collection in the Smithsonian. By tracing some of the history of this collection, I elucidate how recording information validated field data, objects, and individuals within the construct of nineteenth-century science as a metric of knowledge. This collection serves as an example of how “field data” became museum data, how objects became specimens within the museum. I also show how this process was an embodied, performative practice. My analysis of the documentation takes into account materiality – the labels and fields denoted, the space allotments for each piece of information, and the frequency with which

18

Introduction

these standard categories were applied and used. Listing objects in ledger books also allowed museum researchers to keep track of the circulation of objects through specimen exchange, where similar-looking objects were con­­ sidered duplicates and transferred out of the museum.70 Reading these ledgers also exposes tensions between the ideals of data and object collection and the practicalities of the mundane work of record keeping. The large, bound ledger books reveal the limitations of the application of a scientific, natural-historical approach to collecting and recording information about objects, and the ledgers take on another, more fluid, documentary role as pictures and images fill their pages. What I hope readers understand from this chapter is that the affordances and constraints of the media technologies the museum used to record object collections affected, in no small part, how those objects were listed, classified, renamed, and reimagined. These themes continue in the third chapter, “Ordering Devices and Indian Files: Cataloguing Ethnographic Specimens,” which takes a broader view of a later recording technology: the card catalogue. I examine the card catalogue and how it was used in the context of early ethnological practice from the late 1800s to the mid-twentieth century. I look briefly outside the Smithsonian to similar developments across European and other North Amer­ican museums at this time to plot the first tracks of a much larger history in the development of recording material culture. Seen in Europe and in libraries by one of the Smithsonian’s curators, Otis Mason, the card catalogue was brought into the Department of Anthropology as a tool for creating a universal index of all human material culture. I see the development and adoption of the card catalogue as an organizing technology, and suggest that this method and its material affordances worked to stabilize the ideas, forms, and nomenclatures that had been variously and non-systematically applied in ethnological work until this time. This chapter furthers the argument for a practice-oriented model of documentation and media history by analyzing the practice of cataloguing. At various points, cataloguing was characterized as “deadwork,” reflecting its mundane and boring aspects and its practice by largely untrained workers. Once objects came into the museum, they were haphazardly unpacked and sorted, based entirely on the desires of curators (or who was in the building at the time), non-specialist collectors’ notes, and administrative staff (mostly women un­ named in the archives). Those responsible for describing objects often fell victim to normal and uniquely human constraints – boredom and mistake

“The Making of Specimens Eloquent”

19

making. I interrogate this entire process to bring attention to the human ways in which data become a practice, and how this practice is at the heart of much socio-technical phenomena. Chapter 4, “Pragmatic Classification: The Routine Work of Description after 1950,” looks specifically at routine just after mid-century and bridges the long history of the institution with the moment just before records were computerized in the late 1960s. I use the example of a curator at the time, William Sturtevant, who was, like Otis Mason, obsessed with order and chaos in the collections documentation. His attitudes represented an approach to documentation that was changing, as new standards and new methods in anthropology developed. In 1969, Sturtevant wrote a new field guide for collecting ethnographic specimens, which sought to identify “good” specimens in a new context. This guide can still be found in the drawers of the Anthropology Department staff, and it holds its own epistemic loyalties – some of which I disentangle in this chapter. I also investigate the emergence of museum data standards, connecting the legacies of classification with the pragmatic needs of a growing and modernizing bureaucracy. Using information from informal and formal archives and interviews with staff, I try to make evident the work procedures through which museum staff codified material culture during this time. The first computerized inventory systems required standardization to function; yet the adoption of such standardization is far from neutral data management. Computerized systems, like the card catalogues and ledgers before them, were not just simple pragmatic tools for classification: they encoded values, norms, and ideologies from long ago. The fifth chapter, “Object, Specimen, Data: Computerization and the Legacy of Dirty Data,” looks at the development of the computerized index of material culture in the Smithsonian, nearly eighty years after the establishment of the card catalogue had become the primary source of information about the collection. It examines the phenomenon of “bad” or “dirty” data – that is, words, phrases, categories, or descriptions about cultural heritage that do not fit into predefined standards, and that may be outdated, incorrect, racist, or harmful. This chapter tells the story of the practice of documentation, turning the lens on the key individuals at the Smithsonian who were responsible for routinizing the practice of description. This chap­ ter argues that, as computerization increased the importance of standardized nomenclatures in documentation, it also embedded these standardized terms into technologies that are still being used to track, view, and exhibit objects in the NMNH. The continual reproduction of information into

20

Introduction

different media over time often resulted in changes in the amount and accuracy of information. At the same time, information was durable because it was continually replicated. I argue that the catalogue is a performative informational space. From one vantage point, catalogues become the machines of the thought processes: they are the mechanized and routinized material effects of the organization of knowledge. From another, it is possible to see the inherent tension between stability and change, or durability and performativity, as a core characteristic of the history of record keeping at the Smithsonian. This chapter also examines the authority of the catalogue and associated records as bureaucratic documents in the context of repatriation legislation in the United States. In order to repatriate cultural materials from museums, Indigenous communities and museum staff rely on the documentation associated with each object. Every small decision, every revision, and even every mistake in the documentation can have major ramifications for the future of these objects. I demonstrate how the epistemological commitment to pragmatism and dealing “practically” with collections, and the nature of the system itself to impose one kind of ordering logic, has often occluded a more nuanced ethical approach to the management of the information about objects. Thinking about the history of museum documentation allows us to understand the phenomenon of strange, dirty, and bad data – why such data would need to be “cleaned,” and what that need says about the value systems that circulate in bureaucracies built on colonial ideas about human beings and their material culture. The conclusion, “A Museum Data Legacy for the Future,” focuses on the need to be more critical of and attentive to the power relations that be­ come embedded in museum work and that are then crafted by technologies. I reinforce the main claim of this book, that documentation media are not neutral forms but can reinscribe colonial narratives into current practice. Looking within anthropology and museum work, we find a history of bureaucracy and oppression that foregrounds not why intellectual colonialism happens but how it becomes ingrained in institutions.

A Note on Methodology: Researching in an Anthropology Department In November 2013, I was sitting in the main lobby of the National Museum of Natural History in Washington, DC, watching a performance, by a

“The Making of Specimens Eloquent”

21

group of Tlingit dancers, of a Tlingit Killer Whale Clan Crest hat, Kéet S'aaxw, which included the original hat and its almost identical copy. The dancers had been invited to perform their dance because one of these hats – a replica of the original Kéet S'aaxw that was made in the community more than a hundred years ago – was to be put on display with the opening of the Q?rious gallery, a new educational section of the NMNH.71 The replica was made in 2012, well after the original hat had been returned to its community through the repatriation process under the 1989 NMAI act. It was created using a process of laser scanning and computer-guided milling of wood, along with painting and affixing skins and hair,72 and represented a collaboration between the Digitization Program Office, the Repatriation Office, and members of the Tlingit community. This project demonstrated that museums could retain objects in some forms while repatriating the originals to their communities of origin. As the two hats were danced, they moved together and separately through the crowded room. This performance confirmed the necessity for the Tlingit people to have their original hat back and in use; but it also demonstrated the potential value in having a scanned and digitally carved replica for the education of visitors about the role that objects and repatriation play in museums. At the same time, the Smithsonian’s Repatriation Office was working to digitize a group of objects from another Tlingit collection – known as the Hoonah Repatriation Collection – which had been returned to its clan in October 2013.73 The collection of fifty-three shamanic funerary objects had been repatriated, but then lent by the Hoonah Indian Association back to the Smithsonian to be utilized in a unique collaboration. As part of this project, the objects were scanned and then printed using a 3D printer at the Smithsonian Institution Exhibits, a massive warehouse storing and preparing materials for in-house and travelling exhibits. This repatriation project is only one example among hundreds where Indigenous communities, in both Canada and in the United States, have fought hard to retrieve objects and belongings kept in museums all over the world. When I first witnessed how the Kéet S'aaxw and Hoonah collections were digitized, returned, replicated, and reproduced, it made me question how technologies are used to document cultural heritage in particular ways. Such projects, and the use of these technologies, are not always only about returning objects; they are also about exploring new relationships between museums and first peoples globally, where the rights and privileges previously taken by museums ipso facto are themselves returned.

22

Introduction

As the Repatriation Office and the Hoonah Indian Association work toward addressing new ways of documenting and, in some cases, digitally reproducing the Hoonah collection,74 they encountered the legacy data compiled in the early museum registers from the old, and sometimes incorrect, tags accompanying objects and the outdated terminologies used to describe them. The Hoonah collection ranges from expertly carved rattles and large woven hats to small, intricately designed and carved bone fragments. The collection, its repatriation, and future collaborations between the Hoonah Indian Association and the Smithsonian are what made the case compelling and important to consider. The objects were collected in 1884 for the Smithsonian by a US Navy lieutenant, at a time when Alaska was a United States territory and when the collecting and cataloguing practices of the museum had begun to take shape as a formal museum practice. At the time, research was conducted on museum collections under the care of the curator of ethnology Otis Mason, and a small group of staff. Crafted by the paradigms and practices of Mason and his predecessors, this small collection of objects from Alaska was compiled, recorded, named, and labelled as evidence. Reading object histories, such as those in this col­lection, through their associated documentation gives us insights into the contemporary practices of museum registration and cataloguing. These histories begin to explain why, despite the radical shifts in museum practices since 1884, researchers in the Smithsonian today continually encounter problems made durable by documentation practices that are over a century old. I originally came to the Smithsonian’s National Museum of Natural His­­ tory to conduct mostly archival research as I pursued my doctoral work. I was curious about museums and data ethics, particularly about the ways that museums of anthropology or ethnology could use and reuse their existing collections information to repair or redress their colonial histories and repatriate belongings. I wanted to know how online databases structured knowledge, whether it would be possible to design new systems that respected other ways of knowing, and, if so, what that might look like. When I came to the Department of Anthropology to interview staff, observe practices, and sort through cataloguing records, I found that the history of how we come to use particular anthropological definitions and classifications in museum work was incredibly rich, but had yet to be written. And so, I changed course to look at the forms and affordances of historical and contemporary organizational technologies in the museum, a bigger task than I originally imagined.

“The Making of Specimens Eloquent”

23

The Smithsonian Institution includes nineteen museums and galleries (and the National Zoo), and the way collections documentation is handled and administered differs vastly from one museum to another. Still, there are centralized resources to address the concerns of all of these separate entities. The Office of the Chief Information Officer (OCIO) deals specifically with the management and oversight of the information technology net­work and physical infrastructure that support access to collections, the staff intranet, all specific collections databases, web-based media used in the galleries, and the entire strategic plan for information management at the Smithsonian. Each individual museum has its own technology department as well. At the NMNH, this is known as the NMNH Information Technology Office (ITO), which has four branches: Informatics, Oper­ ations, Photography, and Web. Working with the Smithsonian’s OCIO, the NMNH ITO manages local issues at the NMNH and communicates needs and issues to the OCIO. The NMNH has seven different scientific departments: Anthropology, Botany, Entomology, Invertebrate Zoology, Min­eral Sciences, Paleobiology, and Vertebrate Zoology. In addition, there are a var­iety of other offices and departments, including Smithsonian Insti­ tution Exhibits and the Global Genome Initiative. In total, the NMNH holds the records for more than 146 million objects.75 Every department makes use of its own version of the collections database, yet the collections management practices and fields assigned for records for each department vary significantly.76 Many of the scientific departments use global databases for their specimens and research data, in addition to the NMNH’s localized resources. Generally referred to as the Research and Collection Informa­ tion System, the collections’ images and textual records are maintained using the commercial software program of Axiell’s collection management program, EMu.77 In 2012, EMu held over 5.51 million records, and 756,719 digital images were held in the Digital Asset Management System, or DAMS.78 In the anthropology department, EMu was successfully implemented in August 2003, with the migration of more than 500,000 catalogue records from the previous system, INQUIRE. The Department of Anthropology, established as one of the three main departments of the United States National Museum between 1897 and 1898, with William Henry Holmes as head,79 today consists of several distinct divisions (many of which have changed through time, as I will elaborate): Central Services; Collections Management and Conservation; the National

24

Introduction

Anthropological Archives and Human Studies Film Archive; the Powell Library of Anthropology; separate Divisions of Archaeology, Ethnology, and Physical Anthropology (all of which include curators); and the Re­ patriation Office. Collections Management and Conservation manages the collections at the off-site collections storehouse facility, the Museum Support Center (MSC) in Suitland, Maryland. There, the collections (for all other departments as well) are held in large storerooms called pods, and the conservation, registration, and cataloguing of these collections, as well as any overall collections visits, take place there. The department catalogues each object (or small box or lot) with its own number and record. Research on collections is conducted at the MSC; for example, any curator or researcher wishing to understand the history of a specific object can access the original card catalogue file as well as the volumes of ledger books in a small room in the Collections Management Office at the MSC. As a visiting research fellow there, I witnessed staff consistently use these resources as part of their daily tasks, and today these tasks are often completed using the digitized versions online. EMu, the current collections documentation system, is accessible on the local network to each collections staff member, and is administered for the department by a data manager. As EMu was modified for the needs of each specific department, the data manager in the Department of Anthropology is tasked with managing the curatorial and collections staff needs for the use of the system, ensuring the use of standard index terms, adding new terms, and corresponding with the software team. In sum, the care of collections at the Smithsonian is a practice distributed across many different departments and centralized offices, yet the daily work of modifying records and inputting information in the Anthropology Department is done by a small group of collections management staff. Much of the research underpinning this book relies on formal and informal archival research. Months spent in the cataloguing rooms of the Department of Anthropology, the Smithsonian Institution Archives, and the National Anthropological Archives yielded hundreds of documents that show a long institutional history. I also relied on the informal and professional archives of staff, as well as interviews with some staff, to understand this history in the context of recent efforts. The Smithsonian is an overwhelmingly large institution; these archival sources produced a wealth of documents, but there is much more detail to be found. Connecting archival material, scraps of paper, and staff member’s notes and memories was not

“The Making of Specimens Eloquent”

25

an easy task, and, as every researcher does, I brought my own subjectivities to this history. In particular, I was concerned with how knowledge was occluded, and my selection of documents and evidence privileges these ideas. For any future researchers in these archives, I expect there are many more details to be found within.

Museums, Bureaucracy, Colonialism What lies within these pages can be seen as an answer to a question proposed over twenty years ago by Kenneth Dauber, in his essay on bureaucracy and ethnography. He suggests, as I do in these pages, that it is to “seemingly mundane technologies – files, charts, and records – that we should turn in grasping the most durable source of ethnographic authority.”80 As Margaret Bruchac warns of Indigenous informant and museum archives, “Each document is an artifact that informs some part of the larger picture; yet if it is misunderstood or miscatalogued, it could function as a tool of erasure.”81 My goal here is slightly different from Dauber’s, in that I wanted to see not why settler colonialism was able to proliferate, but how. How did these ideas make their way across the country and back into the museum? To me, “how” is a much more interesting question than “why,” and it allows certain possibilities for change. Finding out how something happened points us to questions we would otherwise be unable to answer, without placing blame on simplistic top-down omniscient approaches. It also locates power in the in-between liminal spaces, in relationships, and in “mindless work.” It shows how tools, technologies, materials (whatever you want to call them) were actually used, and how these practices inflicted harm from afar and through time. In this methodology, morality is therefore not located in a particular person’s or institution’s ideology. Instead, it arises from every­day practices and existence. This is a book of specifics. I am interested in little things – the slow movements as terminologies change, the organization of drawers. I am interested in how colonial impulses to name and catalogue became inscribed in the very infrastructure of a museum – the Smithsonian’s NMNH – and what the ramifications of this inscription might be. Although a case study of a single institution, this book stands as an example of what we might learn when we disentangle the specifics of practices that encoded colonial ideologies about race, progress, and material culture into contemporary

26

Introduction

digital records. I hope what readers glean from this study is not only a social history of anthropological record keeping at the NMNH, but also a broader appreciation and desire for inquiry about the cabinets and digital catalogue records that organize, classify, and perform, and the work that is needed to repair the colonial legacies of data around the world.

“The Making of Specimens Eloquent”

27

1

Writing Desiderata

Defining Evidence in the Field

With scarcely an exception, every expedition of any magnitude has received more or less aid from the Smithsonian Institution. This has consisted in the supplying of instructions for making observations and collections in meteorology and natural history, and of information as to particular desiderata; in the preparation, in part, of the meteorological, magnetical, and natural history outfit, including the selection and purchase of the necessary apparatus and instruments; in the nomination and training of persons to fill important positions in the scientific corps; in the reception of the collections made, and their reference to individuals competent to report upon them; and in employing skillful and trained artists to make accurate delineations of the new or unfigured species. Much of the apparatus supplied to the different parties was invented or adapted by the Institution for this special purpose, and used for the first time, with results surpassing the most sanguine expectations. – Annual Report of the Smithsonian Institution, 1854

The scientific study of humankind, as it emerged in the nineteenth cen­ tury in North America, depended on developing collections of objects. Principally, the practice of science and natural history collecting influenced how early ethnologists took objects to be representative of entire cultures

and ways of living. Early ethnologists were often naturalists and brought the ideals and methods of science to material culture collecting in North America. Collecting involved rudimentary attempts to classify or distinguish the objects in the field; certain objects, with properly contextualized information, made for better specimens. During the early decades of the Smithsonian Institution from 1848 until 1870, the epistemic commitments of natural history significantly influenced the collection and documentation of ethnographic material culture. This material culture included the cultural heritage of Indigenous peoples, which was seen as part of the world that could be collected as “natural” ethnographic specimens. These specimens were found “in the field” and required good information, or good data, in order to make them stand out as exemplary specimens useful for research. Documentation was necessary to ensure that the specimens, seen as types of evidence, could be contrasted and compared. Rules were needed to ensure a kind of scientific accuracy in the collection of these materials. To aid in collecting specimens and in the procurement of what were called “desiderata” (desired things), the Smithsonian published lists that collectors and government workers could take into the field. Known as circulars, or collecting or field guides, they were used to solicit information and objects from budding ethnologists and amateur collectors. Many scholars have explored how the language of science was used to describe Indigenous peoples and explain presumed European primacy and progress.1 Throughout the history of settler colonialism, administrative and bureaucratic structures would enable and solidify these interpretations, creating part of an infrastructure of oppression. Like the early catalogues and atlases from scientific pursuits, many museum catalogues resemble or grew out of attempts to categorize the natural world.2 Applying the methods of cataloguing animal remains in natural history museums toward human beings had the effect of positioning Indigenous peoples as a part of a sublime natural world, as opposed to a supposed evolved, civilized one. As Tony Bennett argues, it was not just the language of evolution that was used as an explanatory principle for studying humans and their material culture; an ocularcentric scientific approach developed that made the object the centre of inquiry.3 As others have argued, the primacy of visual language used to record and disseminate knowledge was a key development within the sciences as well.4 Paper tools, including lists, field books, and ledger books, were used in the collection and documentation of material objects and are part of the

Defining Evidence in the Field

29

history of modern textual ethnography.5 The distribution of field guides or collecting lists is not unique to the Smithsonian, or even to American anthropology and museums: there are likely uncovered histories of collection lists in many European museums as well. For example, it is known that, in the late nineteenth century, the British “Notes and Queries” pamphlets were circulated to ethnologists and travellers during fieldwork, and there are even earlier guides that focus on the collection of language types (philology).6 Yet, in the United States early in the nineteenth century, the publication of lists of desired objects informed the conditions of possibility for the development of ethnographic specimens. By the end of the century, these methods shifted, but paper registers and circulars were continually distributed until at least the 1970s. I suggest that, despite different epistemic com­ mitments to collecting and recording objects, the circular as a form was part of the legitimizing power of early anthropology. Analyzing these documents can draw our attention to the mundane yet powerful methods of bureaucracy. Lists that defined what was desired by the collecting institution allowed the lone traveller, enthusiast, or ethnologist to draw on a range of evidences and knowledges in order to, in Kenneth Dauber’s words, “render complexity manageable.”7 As I will argue, tracing the development of collecting guides or circulars draws attention to how the infrastructure of data about material culture in early anthropology developed. The United States National Museum (USNM) (now the National Mu­ seum of Natural History, NMNH) at the Smithsonian is one of the oldest museums in North America, and its Department of Anthropology has produced and fostered research and researchers who have had profound effects on the development of anthropology as a social science. The Smith­ sonian Institution was established by an act of Congress in 1846, and its founding collections were the result of a bequest by James Smithson, a wealthy European collector and scientist, who, in his will, gave the institute its original mandate: “to increase and diffuse knowledge among men.”8 Ori­ ginally, the Smithsonian was without a formal museum, and its first secretary, Joseph Henry, felt that a national museum in Washington would limit its ability to disseminate research to the public.9 Moreover, Henry was aware of the cost of a museum and believed that government funding should be kept separate from the scientific purposes of the Smithsonian to avoid political influence. As Catherine Nichols has observed, Henry was not opposed to the idea of a national museum, but he felt that the Smithsonian should not be the institution to preside over such an important national

30

W r i t i n g D e s i d e r ata

contribution.10 In 1850, Joseph Henry hired Spencer Baird as a curator, and soon after as the assistant secretary of the Smithsonian. Baird brought with him his own small collection, and his own approach to the classification and display of natural historical specimens. Throughout his tenure, Baird was dedicated to maintaining and caring for what he saw to be a growing collection.11 Until 1858, the Smithsonian was without a formal collection of objects, aside from a small bequest, but, in that year, it acquired the collections from the US Government Patent Office, which has been characterized as the first publicly funded collection in the United States, and Congress began to set aside funds for a USNM.12 The Patent Office, organized by the National Institute for the Promotion of Science, contained the early col­lections from the US Exploring Expedition of Lieutenant Charles Wilkes (1838–42). When the Wilkes collection came to the Smithsonian, Baird sought to organize and catalogue it, inspired by his training in the natural sciences. As was common among bureaucratic institutions at the time, Baird adopted a system of museum registration using ledger books to record accessions, modelled on his own work of organizing his personal ornithological collections. The Wilkes collection was then back catalogued: everything that had been collected nearly fifteen years prior was entered into large recording books known as the ledger books. Between 1863 and 1902, much changed at the Smithsonian. The cataloguing department was processing the collections that had been amassed for the 1876 Centennial Exposition in Philadelphia. These collections were accessioned, labelled, and catalogued in the museum’s ledger system. In the mid-1870s, the Indian Bureau of the Interior Department actively engaged ethnologists and surveyors to collect information and objects during fieldwork. Those collections were turned over to the Smithsonian Institution for use in its growing museum collection.13 Smithsonian curators requested that Indian Bureau agents indicate objects that could be obtained by the Smithsonian by physically marking them.14 With the establishment of the Bureau of American Ethnology (BAE) in 1879, which became the professional collecting arm of the Smithsonian,15 and then the institution’s An­ thropology Division in 1883 (later, the Department of Anthropology), the discipline of anthropology became more firmly institutionalized at the Smithsonian. Most significantly, the construction of a new building to house the collection of the USNM was completed 1881 (see Figure 1). By this time, there was an established practice of registering, accessioning, and cataloguing

Defining Evidence in the Field

31

Figure 1   Plans for the United States National Museum building, 1878 | Smithsonian Institution Archives, Image 2002-10710

objects in ledger books. As the Anthropology Division progressively formalized into a department, staff increasingly discussed the utility of cataloguing and arranging objects in the collection. The museum was moved to a new building in 1910; nearly fifty years later, the USNM was subdivided, with the anthropological collections becoming part of a new Museum of Natural History.

Circulating Lists, Circulating Desire The development of anthropology in nineteenth-century North America can be characterized as a concerted effort to craft a serious practice and discipline out of the science of ethnology – the study of living peoples. In the mid-1800s, these studies were mostly in the realm of museums and within the purview of naturalists and geologists. Constructing a proper scientific discipline, however, required the trappings of scientific work: practices of listing, measuring, ordering, and classifying. As James Urry has argued, the collection of ethnographic material was first and foremost a systematic collection of field data.16 In Europe, these were used for the express purpose of enabling travellers or other inexperienced collectors to identify and obtain objects as well as to capture a set of idealized characteristics. At the Smithsonian, circulars, which were published as lists in small booklets or in the Smithsonian Miscellaneous Collections specified the de­ siderata to be collected: objects, vocabularies from different languages, and information.17 While the Smithsonian was actively engaged in collecting objects, ethnologists were aware that collecting everything would be impossible. They relied on non-specialist explorers or citizens to make decisions about what kind of objects were valuable or worth collecting.18 Using lists to prescribe, document, and collect objects, languages, and observations formed the basis for the imagining of a modern ethnographic practice in the mid-twentieth century; yet the history of these documents as media forms or tools is often overlooked. For the early Smithsonian and National Museum, many circulars were issued as guides to procuring biological and geological specimens, but they were soon adapted to collect ethnographic material culture as well. These easily circulated sheets of paper with lists of objects were written by a variety of individuals throughout the history of the institution. By using them, field collectors and early ethnologists across North America were able to collect a wide variety of objects and information, and the Smithsonian was able to extend itself far into the imagined “field.”

Defining Evidence in the Field

33

Some field guides listed the objects that were required for specific expositions and displays. Although these did not restrict haphazard or unsolicited collections from being offered to the museum, they did direct those in the “field” – largely naval or army officers and explorers, and in some cases even the public – to collect certain kinds of objects and information.19 The circulars provided instructions on how to collect and preserve specimens in the field, with notes about what kind of information about them the individual should record. For example, lists of documentary evidence were to be provided, such as where the object was found, its name and use, the collector, and in some cases the tribal affiliation.20 Collectors were instructed to provide specific numbers for the objects as well. Many of the objects that became the anthropological and archaeological collections at the NMNH – what Nancy Parezo has called the “systematic” anthropological collections – were not haphazardly chosen, but were collected because they already fit into a complex system of categorization that prescribed what kinds of objects should be valued, preserved, and named.21

Bodies, Languages, Specimens From 1848 until 1902, the Smithsonian published at least fourteen circulars, lists, and collections guides, many as part of the Smithsonian Miscellaneous Collections series. What is likely the earliest collecting guide was published by Spencer Baird, the first curator of the USNM, in 1850.22 Baird had an explicit interest in natural history, with a particular focus on ornithology.23 He believed that the institution should focus on research and publication, and he was concerned with the formation of collections, seeing such accumulation as the central method by which the institution could achieve the dissemination of knowledge. His influence on the USNM was great for a variety of reasons, but it was his work in creating and defining early museum practices as a curator in the natural sciences that is relevant here – in particular, his application of natural historical principles to the collection of data about people and their objects. Baird’s first circular, “General Directions for Collecting and Preserving Objects of Natural History,” was directed specifically at the collection of what were then considered natural historical collections and was likely drafted before he was officially hired at the Smithsonian.24 The circular is intended for non-specialist collectors and is focused mostly on the proper preparation of “field specimens” – that is, the taxidermy and alcoholic preservation of

34

W r i t i n g D e s i d e r ata

animal bodies. On a small, densely written sheet of paper (which I found attached to an archival record in the institution) are instructions for collecting and preserving any specimens from the field: The Smithsonian Institution being desirous of procuring objects of Nat­ural History for its Museum, would respectfully ask assistance of the friends of science generally, and especially of officers of the Army and Navy. Any specimens of animals, plants, minerals, and fossil remains will be acceptable, particularly such as may be enumerated in the following brief descriptions for their preservation. All contributions will of course be duly credited to their respective donors, in the Museum and the reports of the Insti­tu­tion. By reference to the annexed letter, it will be seen that the Quarter­master’s Department is authorized to receive and forward specimens.25

Next to this, he outlines the specifications for the pure alcohol solutions (“rum or whiskey, the stronger the better”) needed to preserve animal remains. In a section entitled “Special Desiderata,” he describes that he is specifically interested in the “animals and plants of the country west of the Mississippi and Gulf of Mexico.” Yet, alongside the preserved and taxidermied bodies of animals, the guide refers to the skulls of human beings, and singled out American Indians: Skulls of Quadrupeds may be prepared by boiling in water for a few hours ... The flesh can then be removed. Skulls of the American Indians and other human races are desirable when they can be obtained, as also the skulls and skeletons of the large mammal, as deer, bears, wolves, panthers, foxes, beaver, badger, antelope, rocky mountain sheep, mountain goat, seals, buffalo, wild cat, prairie docs, marmots, &c, &c. ... For the purpose of having complete series in the different stages of age and sex, and for supplying other Museums, it is desirable to have a considerable number of the skulls of each species. When possible, at least one skeleton should be procured. It must, however, be remembered, that a single tooth or bone, of an animal, in the absence of anything more, will be of importance. Each specimen should, as far as practicable, have the age, sex, and locality distinctly marked on the bone in pen or pencil.26

It is unclear exactly which “human races” were considered part of an untamed world, alongside buffalo and muskox, yet the remains of American Indians

Defining Evidence in the Field

35

are highlighted here. The list of desiderata provided shows this explicitly as well: “Human races, civilized and uncivilized.” Buffalo Musk ox Mountain sheep or bighorn California wild sheep Mountain goat Antelope Elk Little elk Moose Reindeer or carabou Black tail deer of rocky mountains Black tail deer of the pacific Mule deer White tail deer Deer – other species Beaver Prairie dog Marmots Hares Large wolf, black, white, or grey Lobos wolf Prairie wolf Coyote Indian dog Foxes, all species Sea otter

Common otter Grizzly bear White bear Bears, other species Racoon, especially from California Badger Wolverene or Carajou Fisher Marten Panther Jaguar Ocelot Ounce Tiger cat Wild cat Lynx Civet cat or bassaris Armadillo Peccary, or Mexican hog Walrus, or morse Seals Porpoises Dolphins Whales Manatee, or sea cow Alligator Sharks, stingrees, rays, devil fish; teeth, jaws, and vertebrae.27

Spencer Baird was an avid naturalist, and his collecting instructions are therefore indicative of the kinds of questions that would have been asked about natural historical specimens. The collection of material culture was less emphasized, but the desirability of skulls and other remains shows us that many human beings and their remains were treated as things to be collected and studied within the scientific disciplines, like the remains of

36

W r i t i n g D e s i d e r ata

a marmot or a deer. While many different human remains were likely collected, as I will show, the skulls of Indigenous peoples were of primary interest and were collected alongside, and compared with, the remains of other large mammals, to be preserved for future use, and the only kind of information to be attributed was the “the age, sex, and locality distinctly marked on the bone in pen or pencil.” Without any associated information, the scientist or scholar could apply what were then the standards in scientific method, such as craniometry, directly to the remains and bodies and achieve knowledge through observation. The Smithsonian likely intended that Baird’s guide be used by geologists and naturalists as a way to obtain what he considered to be proper field specimens. The institution’s 1851 Annual Report notes how this circular would have been used as an effective tool for curating the collection of objects in the field: The circular prepared by Professor Baird, describing the method of collecting and preparing specimens, and indicating objects especially desirable, has proved effective in procuring important contributions. Among the objects which should be collected and preserved with care, are the remains of the specimens of the arts of the aboriginal inhabitants of this country, the contents of mounds, and the stone implements found on the surface of the earth.28

Baird’s circular was also received and reviewed by military leaders such as Thomas Jesup.29 Jesup’s correspondence indicates that he received the circular on 25 March 1848, and, in response, he asks Colonel S. Churchill whether specimens he was sending on public wagons could be kept in public stores.30 Spencer Baird’s influence extended to the practices of classification and categorization for the objects that were collected using his guide. An excerpt from an early twentieth-century curator’s report notes the perceived use­ fulness of such field guides for the development of the collections: “In the years between 1843 and 1881 collections reflecting the broadness of [Baird’s] conceptions flowed into the institution in a constant stream, and individuals throughout the United States and other countries were working to build a museum in which everything that has a name should have a place.”31 Baird’s focus was to clearly articulate methods for the collection and preparation of natural history specimens and to indicate objects that were “desirable” for the institution to acquire. Yet his notions of what constituted an ethnographic specimen were not well defined and included objects such as “stone

Defining Evidence in the Field

37

implements” and “industrial products of present tribes.”32 Baird’s circular can be read as a kind of proto-list; it is more akin to a narrative of collecting practice than to the other kinds of lists that followed. It does, however, dem­ onstrate the kinds of questions that were being asked of material culture at the time, and where they were seen to exist and fit within a conceptual, invisible, classificatory system. As the “science” of ethnology developed during the mid-1800s, more and different kinds of objects were sought for collection by the institution. This is reflected in the expanding size of the collection. From 1860 to 1880, the ethnological collections of the USNM more than quadrupled in number.33 There was also a growing desire among ethnologists to collect observations and record information about Indigenous languages.34 In 1852, the institute published Circular no. 1, “Indian Languages of North America,” reflecting the emerging interest in philology in both the American and European ethnological communities. This circular was published for the American Ethnological Society, which was interested in obtaining translations for English and Spanish words in “Indian terms.” It was explicitly looking for origin terms and words that were part of the “primitive stock” of the language, as it felt that borrowed terms were of “no value.”35 The circular included a long list of English and Spanish words, from “God” to “arrow” and from “partridge” to “ugly,” and collectors were to fill in the blanks in the fields. Reading these words should cause us to question the ramifications of this kind of anthropological thinking. We might ask, for example, what was missed when individuals recorded these “translations”? Did the omission of shared vocabularies and borrowed terms ultimately miss the dynamic, cross-cultural existences of Indigenous people at that time? Regard­ less of the linguists’ and ethnologists’ intentions, did the thinking behind the circular reconfirm a notion of the pristine past that did not map onto actual lived experiences?36 In 1861, George Gibbs, an ethnologist affiliated with the Smithsonian, published the guide “Instructions for Archaeological Investigations in the United States” in the annual report for that year.37 This guide was a precursor to his more robust 1863 booklet, “Instructions for Research Relative to the Ethnology and Philology of America,” which served as a document for all field collectors with interests in human culture and languages.38 The 1863 guide was dedicated principally to the collection of skulls and the documentation of Indigenous languages. Gibbs drew explicit distinctions between

38

W r i t i n g D e s i d e r ata

the archaeological and the ethnographic, and the guide was weighted in favour of soliciting information and occasionally objects collected by “officers of the US government, travellers, or residents.”39 His guide is organized into two major sections, on ethnology and philology. Ethnology includes items such as “Crania,” “Specimens of Art, etc.,” and “Hints for Ethnological Inquiry”; Philology includes orthography, vowels, consonants, and comparative vocabulary. The guide is prefaced by a note from Joseph Henry, then secretary of the institution: The Smithsonian Institution is desirous of extending and completing its collections of facts and materials relative to the Ethnology, Archaeology, and Philology of the races of mankind inhabiting, either now or at any previous period, the continent of America, and earnestly solicits the cooperation in this object of all officers of the United States government, and travellers, or residents who may have it in their power to render any assistance. Full credit will always be given for contributions received.40

The very first line of the circular solicits the collection of crania: “Among the first of the desiderata of the Smithsonian Institution, is a full series of the Skulls of American Indians.”41 The horrific ramifications of the collection of Indigenous remains cannot be understated. The circular documents a trend that pervades many of the archival records at this time: the view that Indigenous bodies were part and parcel of the natural world but also of special significance, in that their bodies were particularly desired. What kind of legacy has this left us? It is one, as Jenny Reardon and Kim TallBear propose, that sees Indigenous bodies and even DNA as property operationalized for nationalist capitalist expansion and control.42

“Art, etc.”: Defining Objects and Circulating Reference The leading figures at the Smithsonian considered it important to accumulate as many objects and as much information about as many tribes as possible, and so they made an effort to obtain as many specimens as possible, using some of them to “extend the collection” by means of exchange with other museums.43 Gibbs’s guide lays out three categories into which collectors would slot information and objects: races that had passed away or that might soon be “extinct”; tribes that had disappeared; and existing

Defining Evidence in the Field

39

nations.44 Gibbs specifically noted that collectors had to ascertain the antiquity of the objects so as not to contaminate or confuse them with either more ancient or more modern items.45 Within each of these categories, Gibbs was interested in developing a series of skulls of Indigenous populations, and the guide gives extensive descriptions of how these skulls were to be obtained, and what physical craniometrical characteristics they should exhibit. Just as in natural history specimen collecting, Gibbs saw that the col­ lection of objects alone was a method of inquiry. The collection of material culture was considered part of the work of understanding the totality of culture. In this way, the circular served as a kind of instruction manual on how to collect good information and how to properly inquire about the state and nature of the communities under question. Under the subheading “Specimens of Art, etc.,” Gibbs suggests that everything had value when “considering the completeness of a collection.”46 Yet some items were particularly desirable. Collecting objects provided the material evidence needed to document and comprehend the history of the Indigenous populations that were encountered by researchers, but these populations fell within different classes. The first class, as noted, were “races that have passed away,” and objects associated with these people are categorized as “antiquities.” These include “tools found in northern copper-mines; articles inclosed [sic] in the mounds of Ohio and elsewhere; images common in Kentucky and Tennessee, indicating among other things, the worship of the Phallus,” among others.47 Those in settler society, including the ethnologists at the Smithsonian, often failed to understand that Indigenous communities had a long history of occupation on the land, and so they sometimes mistook old objects as coming from another, extinct civilization.48 In Gibbs’s “third class” – “existing nations” – he listed the following items as important: Dresses and ornaments, bows and arrows, lances, war-clubs, knives, and weapons of all kinds, saddles with their furniture, models of lodges, parflesh packing covers and bags, cradles, mats, baskets of all sorts, gambling implements, models of canoes (as nearly as possible in their true proportions), paddles, fish-hooks and nets, fish-spears and gigs, pottery, pipes, the carvings in wood and stone of the Pacific coast Indians, and the wax and clay models of those of Mexico, tools used in dressing skins and in other manufactures, metates or stone mortars, &c., &c.49

40

W r i t i n g D e s i d e r ata

Gibbs included specific items in the guide for the same reason that Baird had done so earlier: to make certain that Smithsonian collectors would be attentive to the kinds of things the museum desired to have in order to complete its understanding of the world. He also requested that “Indian names” be recorded for each object. The importance of material culture as evidence in constructing this past was paramount. The guide is explicit that, prior to “Western Civilization,” early communities had no written record, and so he considered it the duty of the ethnologist to collect and document everything possible, noting that, “as none of the tribes of this continent, not even the most advanced, ever arrived at the grand and fruitful idea of an alphabetic character for commemorating their thoughts and deeds, almost their entire history previous to the advent of Europeans is left a mysterious blank.”50 In the section “Hints for Ethnological Inquiry,” the focus is on “accumulating data” in the absence of a written history. The collection of information, ranging from tribal affiliation and “Indian names” to the collection of calendars and astron­ omies, was all considered essential data for the “scientific ethnologist.”51 When amassing objects, the “collector” was to include a field catalogue that listed all of the information and observations obtained.52 To validate the information and avoid misinterpretation, having the “independent testimony of more than one” individual was necessary.53 Completed field catalogue categories like “Provenance” became the “positive documentation” of the objects that was required of collectors. For Gibbs, such information led to well-authenticated objects and to “absolute certainty of the true relations” between the object and the community from which it came.54 This authentication depended on the quality of information collected and written down. The guide specifies that each object needed to be assigned a locality (location of origin) along with the name of the collector. Researchers found that the information gathered by collectors was not always accurate, and attributing a locality would enable museum staff to conduct further inquiry into the object to validate the information received in the collector’s field catalogue. Gibbs’s guide asked collectors, when submitting objects, to include information on the name of the tribe and its geographical location as well as the number of individuals in the community. It also laid out the general categories of objects that were of interest to the institution: Picture writing etc. Dress

Food Dwellings

Arts Trades

Defining Evidence in the Field

41

Religion Government Social Life

War Medicine Literature

Calendar/Astronomy History Antiquities.55

The guide instructed collectors to analyze objects and collect information on them based on accepted scientific methodologies of natural history specimens. For example, the guide included the following description of archaeological excavation: When a mound is opened, every particular respecting its position, size, form, and structure, should be noted down on the spot ... and an accurate list should be taken of all the articles found in it. Such as are taken should be properly labelled, and kept by themselves, with the same care that is observed with respect to objects of natural history.56

Classification helped to maintain the scientific accuracy of the data collected, particularly through the creation of a standardized vocabulary.57 However, the Smithsonian immediately recognized that creating a generalized scheme of classification was impossible, and therefore was an impractical way to establish a collection of objects. Gibbs’s guide shows that the kinds of objects desired by the institution had shifted since the publication of Baird’s 1848 circular. In addition, there was an increased emphasis on the collection of information associated with objects, and the beginnings of a system of classification that would separate archaeological materials from “ethnographic” ones. The requirement of good documentation for the purposes of “authentication” was one of the ways that the scientific value of the objects collected was established. The back of Gibbs’s guide contains a list of terms for comparative orthographic collection in Spanish and in English, much like in Circular no. 1. It also includes an appendix entitled “Physical Character of the Indian Race,” which outlines the classification of Native North American people (notably, only “adult males”) via colour of skin and various measurements. A second appendix outlines the collection of terms the tribes used for numbers, with a blank “Table of Numerals” provided so that the field collector could write them down during their work. Lists like these helped stabilize observations about human beings and their belongings, and made them legible as specimens replete with data that could be used.

42

W r i t i n g D e s i d e r ata

In 1867, the Smithsonian published two other circulars: a small pamphlet in Smithsonian Miscellaneous Collections and another longer circular prepared by Joseph Henry. The first, “Suggestions Relative to Objects of Scientific Investigation for the Expedition under Capt. Howard, along the Coast of Russian America,” was prepared by Joseph Henry, George Gibbs, and Spencer Baird. It was directed to an expedition to be sent to Alaska, which the United States was in the process of acquiring from Russia, and Henry was eager to enumerate a list of desired specimens and observations. The circular noted that, among other meteorological, ethnological, and natural history information, the institution desired a “full collection of all the animals known in the Russian American fur trade.”58 Labelling rules were specified to mark each specimen to show the “exact date and locality”: “much of the value of these objects will be lost, unless such facts accompany them.” Informa­ tion on Indigenous languages and grammar was also sought. And, like other circulars, this one also made reference to the collection of human remains: The collection of articles of indigenous manufacture or employment will, of course, form an object of particular interest, and should extend even to the most common and trivial. Drawings, or better still, photographs, should be made of dwellings, tombs, &c. Should a photographer accompany the expedition, it is most important that portraits of good size be taken of individuals of as many tribes as possible, less with the view of displaying their dress and ornaments than their features, form of the head, &c. A collection of skulls representing each tribe should be made as far as possible, particularly of the Eskimo nations, great care being taken to give locality and race.59

The second set of instructions penned by Henry was aimed at the broader public.60 Henry’s “Circular Relating to Collections in Archeology and Eth­ nology” was written primarily to establish the “missing links” in the collection. Although the institution was engaged in arranging the specimens in its collection, its focus was still more on obtaining objects and less on organizing those already in its possession. Thus, Henry requests the help of correspondents and the “friends of science” in “gathering together in the national museum under its charge, a full series of the objects in question as it may be possible to collect at the present day.”61 Henry’s circular includes a list of specific desiderata:

Defining Evidence in the Field

43

Human Crania of both sexes, and in as large number as practicable; the name of the tribe and sex when known, and of the locality and collector to be legibly written on the bone. Photographic Portraits of both sexes of the different tribes. Remains of the tribes now extinct. Among these may be mentioned the axes, arrowheads, pipes, pestles, and other implements of stone found in the superficial soil; pottery, either in fragments or entire, of earthenware, soapstone, or other material; images, and in fact, whatever has been in any way fashioned by the hand of the aboriginal man. The locality of the object should, as far as known, be indicated, as also the name of the tribe formerly inhabiting the region where found.62

He also included lists of objects of “living tribes”: “Various articles of dress used by each sex, and at different seasons of the year; pipes; implements of fishing and hunting, of war, agriculture and manufacture; domestic utensils ...; models of boats, lodges, tents, sleds, etc.; trapping of horses, dogs, and reindeer; mats, baskets, etc.”63 The distinction between the “tribes extinct” and the “living tribes” is reaffirmed, and reinforces the earlier dichotomy between ethnological and archaeological collections in the museum, one that exists to this day. The language in both of these circulars shows a willingness to acquire almost anything – and they both emphasize the importance of associated, contextual information. To call these organizational lists a classification system in the strict sense would be a gross exaggeration. Henry provided not a systematic classification of object types but rather another (similar to the two previous circulars) strategic list of object types that would be of interest to the Smithsonian and would fit the particular kind of collection it wished to grow. Henry, like Gibbs, specified the kinds of information expected to be collected as well. For human crania, the name of the tribe and the sex were the only pieces of information to be recorded. For the “Remains of Tribes extinct,” or archaeological remains, the locality and the name of tribe were sought. In the case of ethnological specimens, more information was required: tribe, locality, date, Indigenous name for the object, “uses of the same, and name of collector.”64 There is little difference between Gibbs’s and Henry’s guides. Both envision the distinction between archaeology and ethnography, and both require similar kinds of information. However, Henry’s circular can be read as a refinement of Gibbs’s less well-defined categories. Both guides stress the need for original documentation. The documentation provided

44

W r i t i n g D e s i d e r ata

by collectors in the field would serve as the foundational authoritative pieces of information for all other publications. In a request to the broad public, in what appears to be a newspaper clipping, Henry Davis (an agent for the American Naturalist and the New Horticulture Monthly) requested on behalf of Joseph Henry “every kind of Indian Relic that is possible to be procured,” including the following: Stone axes, arrow heads, pipes, pestles and other implements of stone. Pottery, either in fragments or entire, of earthen-ware, soap stone or other materials; images, idols, and in fact whatever has been in any way fashioned by the hand of the aboriginal man. Also Indian skulls. The object of collecting these relics is to leave a monument of a race of interesting nations or tribes which has or are fast passing away, and will only be known to the coming generation through these relics. It is of vast importance to the coming races of men that these relics should be collected and preserved in a National Museum. It is well known at the present day that various nations or tribes have existed on this continent, and monuments of their skill and industry are found, but by whom were they made, no one at the present day can tell.65

This strategy to enlist the public in the collection of “Indian relics” and skulls is reminiscent of a kind of proto-citizen science. It also highlights the earlier distinction that, to the ethnologists at the USNM, archaeological antiquities were not connected to current living populations of Indigenous communities. They saw a complete separation between a mysterious group of people who, in the past, had possessed skills to make complex tool assemblages, and the current but “fast passing away” tribes they encountered in their daily work. This distinction is maintained in the way the objects are documented. This approach is a good example of what the anthropologist Mary Douglas identified about classificatory principles: when a thing does not fit within one’s own understanding of order, one constructs a new order through classification, wherein the “outliers” can be explained. Evidence suggests that these circulars had a wide reach. In 1862, a year before the reprinted and expanded version of his guide was published, George Gibbs describes his desire to increase and amend the original 1861 guide, to introduce new classes of language.66 Yet he notes that, because the guide had “already been used so largely,” he was hesitant to change it.67 At the same time, he expressed the desire that the guide be translated into

Defining Evidence in the Field

45

French, because many of the priests working with the Hudson’s Bay Company at the time spoke the language. In 1862, at least one hundred copies were sent to the US Bureau of Indian Affairs.68 The 1861 guide was still in use nearly fifteen years later: by 1876, it had been “distributed widely among the missionaries, Indian agents, travellers, and local collectors in ethnology.”69 The guide had been reprinted several times, and even a new guide based on Gibbs’s earlier iteration was planned in response to increased demand.70 By the late nineteenth century, several more circulars had been produced. In comparison to these later circulars, the earlier ones – Baird’s “General Directions,” Gibbs’s “Instructions for Research,” and Henry’s “Circular Relating to Collections” – hold different kinds of epistemic assumptions and are more closely related to, and perhaps directly influenced by, collect­ ing practices in the natural historical sciences. My examination of the three key circulars produced by Spencer Baird, George Gibbs, and Joseph Henry reveals some of the epistemological frameworks within which the men – and, by extension, the Smithsonian – were working during the early period of the USNM. There is a pattern of continuous improvement or refinement of the collecting guides, and these circulars were revised over time, which might reflect changes in collecting practices. Importantly, the researchers associated with the early institution were part of a global network focused on collecting information and objects. The search for objects was not limited to North American endeavours or collecting trips; international examples of directions for collecting objects existed as well.71 Objects from North America were collected and sent back to Europe – to museums in Berlin, London, Brussels, and elsewhere. Further, it was common practice to exchange specimens across these distances between institutions.72 As objects were exchanged among institutions, practices of documentation became, by the beginning of the twentieth century, a significant topic at international conferences.73

Collecting Information in the Field: 1875 and after Although its collections had existed as part of the Smithsonian for decades, the USNM was officially created only in 1875. Spencer Baird became its curator and assistant to the secretary of the Smithsonian (Joseph Henry), and George Brown Goode was assigned the role of assistant curator of the newly created Division of Ethnology, an entity within the museum. Edward

46

W r i t i n g D e s i d e r ata

Foreman, a museum staff worker, became the person “in charge of the ethnological collection,” and Charles Rau and Otis Mason were made “resident collaborators in Ethnology.”74 Smithsonian staff members were involved in the development of more formalized field methods for archaeology during this time. In 1878, Baird succeeded Henry as the secretary of the Smithsonian, and, a year after that, the Bureau of American Ethnology (BAE) was formed as part of the institution. The establishment of the BAE brought about intensified, formalized studies of North American Indigenous peoples.75 As the discipline of ethnology continued to take shape in North America, collecting for the USNM became a less haphazard and more formalized activity. The BAE was independent from the USNM, but the collections gathered by the field ethnologists of the BAE were all transferred to the Ethnology Department of the National Museum.76 By 1881, George Brown Goode was promoted to assistant director of the USNM under Spencer Baird, and the new USNM building had been completed and opened. Notably, too, the Division of Anthropology was created in 1883 as a separate entity within the Smithsonian. These changes all represent a growing and diversifying institution and department of anthropology. Paired with changes and with increasing numbers of staff, more formalized methodologies developed across disciplines, especially with respect to administrative procedures. At least nine more circulars directing the field collection of archaeological and ethnological materials were published after the creation of the USNM. The first of these, entitled Ethnological Directions Relative to the Indian Tribes of the United States, was published by Otis Mason in 1875 during his time as an unpaid collaborator.77 The Smithsonian distributed circulars to ask “correspondents of the institution to aid in collecting specimens, and to give information as to the existence of special collections, from which unique specimens could be borrowed for copying in plaster.”78 Mason’s pamphlet was distributed widely, and the Smithsonian’s annual report accurately described it as a “systematic schedule” for collecting.79 Mason’s pamphlet was still in circulation in 1878: over 1,000 copies were distributed that year. The Smithsonian’s annual report for that year notes that “it is hardly too much to say that this circular has been scattered broadcast over the land.”80 The guide was distributed to organized establishments of literary, educational, and “scientific character,” to newspapers and postmasters, and by companies, through their agents, to individuals whom they knew to be especially interested in the subject. A copy was mailed with each

Defining Evidence in the Field

47

written communication sent out by the institution, no matter what the subject. The collections that came into the museum based on the recommendations in the pamphlet made the connection between collector and object paramount – more so than earlier examples. This was done to ensure that the information might at some point be verified and to give “due credit” to the collector in the documentation. Such crediting brought prestige to the collector and encouraged more collections.81 Mason’s guide contained by far the most numerous and extensive classification of object types of desiderata. The annual report for 1875 notes the importance of Ethnological Directions and its wide reach at the time: For the better presentation of the subject to collectors and correspondents. Prof. O.T. Mason, of Columbian University, was requested by you [Joseph Henry] to draw up a systematic schedule of the various articles of clothing, ornaments, household utensils, implements of agriculture, weapons of war and the chase, tools of trade, the apparatus used for the pursuit and capture of game, &c., and a pamphlet was accordingly prepared by this gentleman, embracing over six hundred subjects. Copies of this pamphlet were then sent by the Indian Bureau to all its agents, and by the Smithsonian Institution to its correspondents, with the request that they would indicate by a mark on the list the articles that could be obtained, and return them to Washington. This was done to a considerable extent, and authority was given in many cases to proceed in making the collection. Several gentlemen of much experience in ethnological researches were also employed by the Bureau to secure complete collections from the tribes within their reach.82

During the year when Mason’s pamphlet was published, the museum was preparing for the 1876 Centennial Exposition in Philadelphia.83 Thus, the preface to Ethnological Directions directs attention to “objects and investigations which are desirable for the purpose of representing at the Centennial Ex­position the history and culture among the aborigines of America, including the tribes now in existence, and those which are nearly or quite extinct.”84 These materials were classified as four different kinds: specimens now or formerly in use; models in true proportion; photographs and drawings of various kinds; and manuscripts and descriptions of objects, customs, institutions of society, laws, beliefs, and forms of worship.85 Mason clarifies that the information sought by these collections relate to the “tribes of men, their

48

W r i t i n g D e s i d e r ata

natural and social surroundings and the history and condition of culture among them.” The preface divides the “Classes of Ethnological Material” into two main categories: “Man, Surroundings or Environment” and “Cul­ ture.” The “culture” class has seventeen different subheadings of material types, which themselves contain nearly 750 different sub-classes. “Culture” subheadings are organized into functional groups, such as “means of subsistence,” “habitations,” “vessels and utensils,” “implements of general use,” and “means of locomotion.”86 Object categories were designed from a functional taxonomic perspective, envisioned as a way to enable the museum to collect objects from all parts of the globe by using predefined categories or standards of collection. Mason’s ultimate goal was to design a “classified report” of all the material in line with these kinds of ethnological categories, which he defines as “function, geographical distribution, degree of elaboration, material, and classes of investigation.”87 As Catherine Nichols has argued, these categories were part of Mason’s collecting bias, and were situated in historical natural history approaches to understanding Indigenous cultures.88 Mason’s personal interests – he was particularly interested in throwing sticks – contributed to the organization of the guide as well. For example, it contains eight different sub-classifications of hand-held weaponry items and only two sub-classifications of “art.” The guide characterized all objects as equally relevant – a development at odds with earlier circulars. Indeed, the act of deciding what was relevant and what was not was to be left entirely to the “professionals”: “Inasmuch as it is designed to present savage life and condition in all grades and places, care should be taken not to exclude specimens because they are rude or homely; not to collect with a view to artistic effect merely; and not to rely too much on one’s own judgment as to what things are desirable for ethnological study.”89 Scrupulous adherence to truth was highly valued, and the guide recommends attaching an “ineffaceable number” and manuscript cata­ logue with the collector’s name and address.90 Similarities in broad fields of inquiry are evident in Gibbs’s 1861/63 guide and Mason’s Ethnological Directions. In both cases, objects and information considered suitable for collection were listed within the text, or as accompanying tables (as with Mason’s guide), and they show an emphasis on scientific rigour and acquiring positive documentation for the purposes of verification and contextualization. Travellers were expected to acquire “good” documentation, thus ensuring that the objects made “good” specimens.

Defining Evidence in the Field

49

Ethnological Directions presents a more specific classification scheme where specimens in use, models, photographs or drawings, and manuscripts (descriptions of objects, customs, laws) were considered the useful collections to acquire. The “Native and vulgar” name was required by the guide, as was the locality, tribe, use, date of collection, and the number of pieces. “In short,” Mason wrote, “the full history of the object in as few words as possible.”91 In this pamphlet, we see the origins of a standard museum practice of collecting and registration for ethnology, not just a reconfiguring of natural historical practice.92 Mason’s was not just a guide for specifying how and when labels should be written upon objects and belongings; it is also a proto-classification for many ethnological objects that were already in, or would enter, the museum. Establishing the association to collector by marking this on the object (for example) created a permanent link between object and donor, and allowed Smithsonian researchers access to the provenance of the object through its collector as the authoritative source. Other circulars focused on language and archaeology. In 1877, for example, Wesley Powell published Introduction to the Study of Indian Languages: With Words, Phrases, and Sentences to Be Collected, a book that defined the aspects of languages to be collected. The first page of the volume is a blank form with four categories – “Tribe,” “Locality,” “Recorded by,” and “Date of record” – prefaced by the caution, “Be Particular in Filling this Blank.” This form, with its fill-in-the-blank categories, would be used to collect, map, and “preserve” languages from all over North America. Powell’s guide was explicitly a revised version of Gibbs’s 1863 circular, which Powell says had greatly stimulated investigation, giving wiser direction to inquiry, and the results have abundantly proved the value of the “Instructions” and the wisdom of its publication; and it serves to mark an epoch in the history of ethnographic investigation in America. The material which has thus been accumulated is of great amount, and its study has led to such important conclusions that it is deemed wise to prepare a new system of instruction, more comprehensive in plan and more elaborate in detail.93

Powell’s volume includes a detailed description of why the alphabet is important to collect. It also includes a list of categories in which Indigenous words should be collected, with blank spaces left in which the collector could transcribe terms:

50

W r i t i n g D e s i d e r ata

I. Persons II. Parts of the body III. Relationships IV. Social organization V. Governmental organization VI. Religion VII. Disposal of the dead VIII. Dress and ornaments IX. Dwellings X. Implements and utensils XI. Food XII. Games and sports XIII. Animals

XIV. Trees, shrubs, fruit, etc. XV. The firmament, meteor­ologic, and other physical phenomena and objects XVI. Geographical terms XVII. Geographic names XVIII. Colors XIX. Numerals XX. Measures XXI. Division of time XXII. Standards of value XXIII. New words XXIV. Phrases and sentences.94

In total, the guide is 104 pages, containing mostly blank pages where lists of words could be added, translated, and “collected.” These forms, whether used by the collectors or not, give us insight into the kinds of information or data seen as valuable at this time. In 1878, Joseph Henry (likely supported by Otis Mason) published Circular no. 316, “In Reference to American Archaeology,”95 a revision of an 1867 circular. The updated guide solicited an increased amount of information and larger number of collections from those to whom it was distributed while focusing on encouraging collectors to conduct correct and scientific archaeological excavations.96 It directed that professional collectors or amateur archaeologists properly describe each mound or site, starting with the locality, which the guide described as the “county, township and the distance and direction from the nearest post office or railway station.”97 In addition, collectors were to specify the number and manner of groupings, the shape and size of the site, the “internal structure,” and the contents. With respect to contents, the circular established as desiderata earthenworks and human remains found within mounds or other excavation sites, but objects of every kind were to be collected. Even “mundane” objects had scientific value, and accurate recordings were desired of “whatever is observed.” Essentially, the Smithsonian was trying to compile the most complete and accurate record of all North American archaeological finds at the time, publish a directory of every archaeological collection, and, by doing so, build up a complete history of the continent.98 This circular again provides key

Defining Evidence in the Field

51

evidence that the remains of Indigenous peoples were being actively sought, collected, and preserved, along with associated objects: The most desirable objects among the contents of these mounds and other earthworks are the human remains. In order to preserve the crania and bones, as well as bone implements or other friable objects, they should be covered with boiled oil, or with a weak solution of warm white glue. When this has dried sufficiently, the process may be repeated until the tissue is completely hardened. Small fragments may be immersed in the liquid at once ... [A collector] should disregard no object however insignificant it may seem, and record with the most scrupulous accuracy whatever is observed.99

As in the past, accuracy was facilitated by having the individual who collected the object forever associated with it. Each specimen was labelled with the donor name so that, “wherever the object may be, the source whence derived will be known.”100 A list of questions was included in the pamphlet so that everyone who received it could write as much as possible about what they would send back to the Smithsonian; in some cases, collectors also provided sketches. There were many responses to Circular 316, but it is difficult to tell which exact collections were amassed as a result of this effort. Otis Mason summarized the correspondence that was elicited by the circular, but this record amounts only to remarks and descriptions of the kinds of finds and excavations that were taking place across the country at this time.101 One collector, James Swan, certainly had a copy of the guide and responded with information on, as well as a collection of, over a thousand objects, which were sent to the Division of Anthropology. Over 100 responses were published in the 1879 annual report, which included lists of information provided by individuals, many of which relate to burial practices and archaeological land features. This kind of citizen science made it possible for the museum to pinpoint exactly where interesting sites were, and the responses to Circular 316 demonstrate the ways in which the ideals of the core extend to peripheries – that is, how the bureaucracy of the institution manifested itself in the backyards of people all over the country: This circular has proved more prolific of correspondence than was anticipated by its most sanguine friends. Inquiry soon followed inquiry for more detailed

52

W r i t i n g D e s i d e r ata

information as to desiderata in the way of specimens and information; requests were continuous to know if this or that article would be welcome; offers to lend objects for copying were numerous from individuals possessing unique and choice specimens, valuable to them only as heirlooms or curiosities, but who, while easily convinced of the little value of isolated collections in comparison with the importance to science of one grand, complete series, could not be persuaded to part with their archaeological treasures; now and then private cabinets of more than ordinary character and extent have been brought to light; and, finally, modern “manufactories” of relics have been detected, and in this way forgeries, to some extent, driven from the market. Thus an extra amount of correspondence has been entailed that has proved no inconsiderable addition to the previous arduous labors of the officers of the Institution.102

Between 1880 and 1889 three more circulars were published: Charles Rau’s 1883 “Circular Relative to the Contribution of Aboriginal Antiquities to the United States National Museum”; Cyrus Thomas’s 1884 “Directions for Mound Exploration,” and the Canadian Institute’s “Circular” (1889).103 Charles Rau was the curator of the Antiquities Department at the USNM, and his circular begins with a complaint that objects coming into the museum were often without good and “accurate” information. He lists the kind of information that should accompany collections found at particular sites, such as shell-heaps, graves, and mounds. “The Explorer,” Rau begins, “it hardly needs to be said, should collect the human remains and all articles found within them.”104 Thomas’s short circular, which was a series of directions on how to properly explore mounds, shows the development of what would become archaeological field investigations in the United States. He specifies everything from the kinds of measurements needed to the kinds of drawings that should accompany these observations. The Canadian Institute circular is more akin to earlier Smithsonian circulars; however, it is less focused on the collection of objects and more on the recording of general observances such as tribal organizational structure and philology. In 1902, William Henry Holmes and Otis Mason published the guide “Instructions to Collectors of Historical and Anthropological Collec­tions,” which is a revision of Mason’s earlier Ethnological Directions that is directed specifically at expeditions to the Philippines.105 This guide makes it clear that government agencies were still responsible for collecting objects and were

Defining Evidence in the Field

53

seen as the agencies best equipped for “additions to anthropological treasures.”106 The guide documented the desire of the USNM for two “principal classes of objects” – natural history specimens and those that were categorized as “anthropology” (i.e., related to the more “primitive tribes”). The competitive collection of specimens had become particularly important because other museums were quickly amassing collections: Holmes and Mason note that an opportunity would be missed in gathering material heritage if others got there first.107 This circular distinguishes between historical and anthropological collections (the latter being connected with the scientific study of the human body). Where the “historical collections” focused primarily on weapons and munitions that were of particular interest to Otis Mason, the anthropological collections were classified into two types, somatological and ethnological. Somatological collections were considered to pertain to “man’s body,” while ethnography was concerned with the “works of man.” Both could be documented by objects, photographs, or descriptions: “Anthropological collections are of three kinds, to wit: Objects or specimens; photographs and drawings; narratives and descriptions. The specimens form the cabinet, the pictures give life to the specimens and show them in their true environment, the descriptions form the basis of all labels and of the literature of anthropology.”108 Mason and Holmes requested that somatological data be collected from Filipino populations in particular, arguing that “in [their] veins runs the blood of all the varieties of mankind.” These observations were to be of people and their physical attributes, as well as anatomical collections of “crania, pelves, long bones, whole skeletons, abnormal bones, results of rude surgery, trephining, head flattening, or deformations due to design or undersigned custom.”109 The connection between collection of material objects and colonization is literal and direct: “Physiological Processes: Including vital processes in health and disease, and physiological psychology. The former has reference to race adaptability to climate, immunity from disease, and all such idiosyncrasies as will be useful in colonization.”110 The reference to the medical sciences to justify the collection not only of measurements but of entire bodies is not new, yet in this document we see the close connection between science, racism, classification, colonialism, and early anthropology. The “ethnological collections” consisted of a list of twenty-three different types of objects that were thought to be appropriate for collecting. Collectors

54

W r i t i n g D e s i d e r ata

were to assign each specimen with a tribal affiliation described in detail, including whether such affiliation was “blood” related or of a political nature. The guide emphasized that attention to detail in the cataloguing and labelling of objects was of utmost importance: Everything depends upon the label. At least the use of the object should be told. The common shipping tags are good enough for all purposes. Each object should bear a collector’s number and this should correspond with a number in the collector’s notebook. On the tag should be given the common and the native name of the object, the locality, the tribe, the use, and the collector. Much more will be written in the notebook.111

Holmes and Mason note that the museum wanted reproductions and models, which were used as teaching tools, but it also requested that all of the tools used to make objects be collected as well – evidence of a museum explicitly attempting to collect process. Ethnological specimens and inquiries were further classed into the categories of arts and activities, social life, aesthetic culture, knowledge, religion, and environment. Under the “Arts and Activities” section, there are twenty-five classes or “object types”: food drinks narcotics, drugs, and medicines dress and adornment accessories to dress habitation and other buildings furniture mechanics’ tools primitive engineering machinery stone working ceramic art metallurgy

woodcraft textile industries agriculture milling arts hunting fishing animal products curing fish paints and dyes transportation on land transportation by water metrics and commerce.112

Table 1 lists the other five sections in the “ethnological collections” along with the object types for which observations and information were to be collected.

Defining Evidence in the Field

55

Table 1

List of observances and object types desired Social life

the family political organization army and navy lawmaking courts of justice administration customs and unions

Aesthetic culture

fine costume music graphic arts sculpture and carving ceramic art textile art art in metal gardening formalities dramas oratory literature

Knowledge

traditions sayings knowledge philosophy the pantheon

Religion

priesthood sacred places worship religious literature celestial environment

Environment

geologic environment meteorologic environment mineralogical environment botanic environment zoologic environment

Source:  William H. Holmes and Otis T. Mason, “Instructions to Collect­ ors of Historical and Anthropological Specimens: Especially Designed for Collectors in the Insular Possessions of the United States,” Part Q of Bulletin of the United States National Museum, no. 39 (1902).

56

W r i t i n g D e s i d e r ata

Creating these lists was an attempt to structure and standardize the field collections and observations, but classing observances was also a way of establishing which fields of information were more valued than others. The categories here reflect Mason’s interest in compiling typological relationships between objects from places around the world. The epistemological and ontological questions posed in these guides present a fascinating account of how particular (and new) brands of “science” – anthropology and ethnology – were developing in the second half of the nineteenth century.

Specimens, Documentation, and Value Scholars have argued that the collection of objects from the years 1875 to 1920 played a significant role in the professionalization of anthropology as a discipline, and that the collection and recording of objects was not peripheral to the documentation of culture itself.113 Two of the Smithsonian circulars examined here – Gibbs’s (1861/63) and Mason’s (1875) – are characterized by common traits of anthropological inquiry at the time. The 1863 guide was designed to elicit useful information, and occasionally objects, collected by “officers of the US government, travellers, or residents.”114 In contrast, Holmes and Mason’s 1902 guide focuses on collecting the objects themselves, and there is a shift in the way that knowledge is characterized. However, good documentation is consistently considered to be the most significant aspect of all collecting. Once an object was identified for collection, its scientific value – or its value as a specimen – was established and verified in two ways: first by the collection of multiple specimens, and second by the collection and recording of accurate and relevant documentation.115 Multiple specimens were often collected so that they might be distributed to other museums after their return to the Smithsonian and validated against other collections of similar objects. Documentation also served a second function of “extending the collection by exchange,”116 and good documentation made it easier to determine which objects were the “same” and available to be exchanged. This practice of exchange was a common occurrence for all collections at the time, not just ethnological ones.117 These kinds of ethnological inquiries occurred in an effort to overcome what anthropologists at the time perceived as the absence of history in the Indigenous communities of North America. In the absence of written language, Indigenous peoples were deemed to have no history and the collection of objects and their associated information crafted an “object literacy” out

Defining Evidence in the Field

57

of which anthropologists believed they could create a history. Using “natural historical” observations and methods enabled early anthropologists to formally critique and understand these objects. These methods can be seen as the origins of the formalist approach to understanding culture based on visual attributes of technologies and things, and a hallmark of a functionalist approach that classifies together all objects that have the same role. Part of collecting “good” ethnographic knowledge and specimens meant establishing a way to make the information legible, transferable, and verifiable. The early field guides pay particular attention to developing a standard and classified system of documentation with respect to knowledge and material heritage through the development of a recording system for field observations and catalogues. These guides were for the collection not only of objects, but also a range of information relating to the local populations. The collecting guides demonstrate that paying attention to the “pure facts” was paramount. Personal interpretation and the “feelings of a low grade of culture” were to be avoided, as they did not add to the knowledge of a scientific discipline, something of high importance.118 The writers of these guides considered Native thought to be unreliable and untrustworthy; thus, they stressed the importance of the collector’s trying to include independent testimony of more than one individual. As the 1863 guide noted: “The character of the Indian mind is so essentially different from that of the white man, they think in so different a manner, that many precautions are necessary to avoid giving them wrong impressions of our meaning, and of course obtaining incorrect replies.”119 There was no true interest in a full understanding of Indigenous knowledge in the object’s documentation; rather, the hope was that the collection of information itself might give in­ sight into the “mental idiosyncrasy of the people.”120 In order to validate “good” documentation, thorough examination was recommended. The guides recommended fully understanding the environment in which the object was found, with particular focus on objects that had been excavated or found in mounds or caves.121 Good documentation, as recommended in Mason’s 1875 guide, required a clear connection between the collector and the object. Professionals often considered novice collectors untrustworthy sources of information. Mason recommends that field collectors should be wary about relying too heavily on their own judgment, as he assumed that the collectors did not have the “scrupulous adherence to truth” that was required of Smithsonian scientific researchers.122 Yet, by naming objects carefully and methodically, collectors could ensure that their data

58

W r i t i n g D e s i d e r ata

was objective and accurate. For example, the guides specify giving correct names to the “tribes,” which an object comes from, which were validated by conferring with neighbouring tribes.123 These field guides can be characterized by an emphasis on scientific validation, and the classification of objects in situ, at the site of collection where the information was easily attainable. Within the paradigm of sal­ vage anthropology, the guides assumed that the populations from which these objects had been obtained would soon be extinct and, therefore, that museum efforts toward collection benefited the tribes themselves. However, as Margaret Bruchac has fruitfully pointed out – such beliefs reveal only one side of the story, and an archivally biased one at that.124 Yet these perspectives were inscribed on institutional documentation and made durable in the museum’s ledger books and catalogue records. One source of an object’s value was the information or data associated with it. Key pieces of information, which were to be recorded at the site of collection, gave value to the objects as specimens, even ones that were not immediately recognizable as valuable. As Otis Mason himself stated, “under the secure guidance of things well authenticated, even rubbish will become useful.”125 Nancy Parezo notes that the Smithsonian’s emphasis on systematic documentation was a rule for researchers and collectors and functioned as a standard by which collections were amassed. She argues that these documentation practices were as valuable as the objects, and that the objects with good documentation increased in scientific value.126 As I have noted above, all the early field guides emphasized the necessity of recording an object’s “locality” and “collector,” insisting that the locality was one of the most important pieces of information associated with specimens. Notably, when the hunt for relics became a business in the late nineteenth century, vandals were extremely careful to create catalogues and provide locality information for collections because they knew that doing so would increase the value of the specimen – even in the absence of the original information or any contextual information at all. As noted in the Smith­sonian’s annual report for 1901: Various governmental explorations called widespread attention to the ruined pueblos of the Southwest, and soon it was found that relics from these pueblos had commercial value. With this entering wedge, the collecting of “relics” became a business, and men traversed the region for the sole purpose of tearing up the ruins for their private gains. Almost every trader either employed

Defining Evidence in the Field

59

Indians to dig or bought all the specimens that Indians brought in at a nominal price, and many were the men who had “collections” for sale. A few of these individuals, profiting by the scientific methods of governmental and institutional explorations, were careful to catalogue and localize the specimens as far as possible at second hand, finding that such data increased the value. To give an idea of the extent of this vandalism and unscientific collection, it may be said that from one town alone during the past ten years about 20000 specimens have been shipped; from other neighbouring towns about 7000 specimens. From the same points during this period about 10000 specimens have been shipped by scientific exploring parties. The speculative collecting was from Indian reservations, railroad and government lands.127

These observations imply that objects were considered of value to museums only when they came with data that, at the very least, appeared in a form that was considered truthful or verifiable – in this case, a catalogue book. The collecting guides of the late nineteenth and early twentieth century all represent diversified attempts at amassing collections. Guides were crafted for specific areas – archaeology or linguistics – and for specific areas or time periods, as in the case of the 1902 “Instructions for Collectors of Historical and Anthropological Materials,” which was created for those who collected objects while visiting the “insular possessions of the United States,” specifically during the Philippine-American War. Circulating lists and forms for all types of observations transmitted meaning and extended the Smithsonian’s reach into the far corners of the world. Inasmuch as the objects themselves transmitted meaning, the circulars and field guides were devices used in the development of networks, akin to a kind of interessement device.128 Seeing circulars as tools and devices helps us to understand how these guides were able to solicit objects from the field, and how the Smithsonian extended its influence within other collecting situations. These circulars can be (and were) interpreted in different contexts, which served to stabilize meanings that were dispersed through time and space, across the world, and spread American scientific ideologies. Latour’s concept of “immutable mobiles” has relevance for the analysis of the Smithsonian’s circulars. He uses the example of mapping new territories. A cartographer maps a landscape that does not move and that remains the same, but, in doing so, creates inscriptions of the landscape that can be brought back and accumulated with other inscriptions. The map is the immutable mobile, an object and inscription device that both represents an

60

W r i t i n g D e s i d e r ata

existing thing and carries it into other contexts. Scientific diagrams are also good examples, as are laboratories or exploring ships. An exploring ship, as Latour details, brings its own ideologies, systems of thought, and the sensibilities of its passengers to other places.129 As Latour has argued, these kinds of immutable mobiles were responsible for the spread of European science and market capitalism.130 This analysis can be extended here to the topic of anthropological collecting. The process of creating immutable mobiles also requires the creation of visual inscriptions, particularly in the sciences. As Latour elaborates, “We are so used to this world of print and images, that we can hardly think of what it is to know something without indexes, bibliographies, dictionaries, papers with references, tables, columns, photographs, peaks, spots, bands.”131 Anthropo­logical field guides acted as immutable mobiles: the lists for collecting prescribed the collections that could be located in the field and outsourced a kind of local classificatory scheme no matter where the collector journeyed with this information. The circulars, used on separate expeditions and by different people and in different contexts, maintained the museum’s immutable qualities but made the act of collecting mobile. In this way, they were able to “act a distance.”132 The ability to circulate in various networks but retain the same configuration and shape is a hallmark of these guides, which (in their desired and perfect shape) could be reproduced without alteration. Just as in the construction of maps in scientific expeditions, the curators followed standard conventions, and made objects and individuals encountered legible to other travellers and collectors, despite the fact that each guide was drafted for different places and times. This immutable mobility allowed objects to be read across cultural areas but to represent new knowledge. It also meant that the directors of the institution and the curators held considerable influence over the way in which the collections developed during this time.

Traces of Resistance Stories of how collectors used or relied on documents to understand what was desired by the Smithsonian do not, of course, accurately represent what actually happened “on the ground” during the negotiated process of ethnographic collecting. Focusing on these documents, as I have done, allows for a richer understanding of the role of documents at a time when academic disciplines – particularly those concerned with understanding human beings – were affected by rigorous bureaucratic practices that were a key part of

Defining Evidence in the Field

61

institutional functioning. But such a focus elides any Indigenous community’s experiences, practices, and encounters, as well as the ways in which the people who were seen as the “site” of collection actively participated in and resisted this role. A fuller investigation of field notes and community histories would reveal the multitudinous ways in which privilege and power were practised in localized places rather than being told from the perspective of traces of ethnographic collecting materials. Nonetheless, we can read the circulars to find moments of change, break, or disjuncture, which can allow us to see the acts of resistance to the imposition of colonial collecting. Circular no. 1, “Indian Languages of North America” (1852), makes it clear that simply walking into a community and requesting (or taking) objects and writing down linguistic traits would not be possible, and points out the need for “Indian Agents and Interpreters” and the need to hear the words from the most “Intelligent Natives.” Read one way, these directions speak to shifting power relationships, where Indigenous people were needed by the officers of the Smithsonian in order to validate their information and field data – a need that complicates the simple give-receive narrative often expressed in scholarship. In another example, in Holmes and Mason’s 1902 guide, “collections ... are not so difficult to obtain and are secured through charming personal friendships.”133 Likewise, in George Gibbs’s 1863 philological circular “Instructions for Re­ search Relative to the Ethnology and Philology of America,” one section, which addresses the resistance of Indigenous communities to the gathering of human remains, speaks loudly: The jealousy with which they guard the remains of their friends renders such a collection in most cases a difficult task, but there are others in which these objects can be procured without offence. Numerous tribes have become extinct, or have removed from their former abodes; the victims of war are often left where they fall; and the bones of the friendless and of slaves are neglected. Where, without offence to the living, acquisitions of this kind can be made, they will be gladly received as an important contribution to our knowledge of the race.134

As framed by Gibbs, Indigenous agency and protection becomes “jealousy,” and individuals are cast as having childlike emotions that serve only to upset paternalistic collectors. Traces like these remind us that Indigenous people were not simply passive recipients of colonial processes, but were actively

62

W r i t i n g D e s i d e r ata

engaged in the negotiation of power. Terms like “jealousy” and “offence,” just like Holmes and Mason’s “charming friendships,” are colonial coded words used to delegitimize the power of entire communities – a strategy that is still used. In the attempt to apply rules of description in order to render ethnological collecting as scientific research, and through establishing a practice of reproduction for ethnographic objects, the Smithsonian stabilized the con­ ditions for a “proper” ethnographic object. This process of identifying “desideratum” was a key part in the formation of anthropological knowledge and research at the institution, and it framed Indigenous knowledges in important ways that continue to haunt collections practice today. The criteria for establishing the value of ethnographic objects were not perfect, and, like any system of verification or categorization, produced variant specimens, including ones that did not fit within the idealized categories for collection. However, strategies for validating objects and for determining their suitability as “objects of knowledge,” and the administrative practices that developed out of them, became normalized categories for ethnographers and anthropologists. Objects collected in the field became objects that held scientific value, to be measured as specimens. Guides to collecting objects in the field and the system of ledger books were both mechanisms that en­ abled collectors in the field and museum workers to slot objects into existing categories of knowledge that they were then able to measure against existing anthropological inquiry. These information strategies for collecting objects and cataloguing information were situated within ideals of the natural sciences. What is most interesting is that the circulars often prescribed the collection of the names, linguistic affiliations, tribal designation, and even time of use for particular objects. This information was to be collected and noted in the field and delivered along with the objects to the Smithsonian – but it was rarely recorded as part of the “object record” in the ledger books once they came into the USNM. Why was this? Perhaps it was in part because collectors simply were not providing the kinds of information that was desired, and the individuals responsible for the record keeping of the collection often lacked information about the objects themselves. But we can also look to the material form of the ledgers. These were stable documents with defined fields, and certain collected information simply did not meet the space or contextual requirements or incorporate diverse kinds of knowledges about each individual object. In order to understand how the objects

Defining Evidence in the Field

63

and the associated information collected in the field was made stable or durable then, we must understand the actual process of record keeping at the institution, beginning with the use of the large ledger books. The next chapter looks at how epistemic assumptions were present in the formation of the ethnographic catalogue as a whole, and how these assumptions came to be inscribed upon this other museum media.

64

W r i t i n g D e s i d e r ata

2

On the Margins

Paper Systems of Classification

Writing is mnemonic, the history of communication tells us; it is preservative. And so are printing and bookmaking: each ... [type of book] formed a class or category of blank because each catered to the repetition of certain kinds of writing. If writing is preservative, these books preserved preservation. Their design, manufacture, and adoption worked to conserve patterns of inscription and expression. A blank blotter catered to the repetition of inked inscription only – no matter what was written or drawn – but most blank books would have worked however modestly to mold, to direct and delimit expression. Order and invoice books, for instance, like ledgers and daybooks, catered to inscriptions accreted according to the vernacular habits of trade and the long-standing formulas of accountancy. – Lisa Gitelman, Paper Knowledge

After individuals in the field collected, amassed, and exchanged objects, using the Smithsonian’s guides discussed in the preceding chapter, they sent these objects to the institution to be formally accessioned and catalogued in the institution, a practice that began as early as the 1850s.1 Systems of formal documentation inscribe and re-inscribe many of the epistemologies developed in the forms that come before them, so that the reproduction of names from the circulars becomes a recurring pattern, even as information

moves to new documentation media. This is something common to all sys­ tems of bureaucratic record keeping; yet it takes on special importance in the context of memory institutions that were part of broader colonial projects. In this case, these data legacies, as I have termed them, impact our current conditions of possibility for making claims for the return and restitution of cultural heritage. One of the key technologies for sustaining information through time are ledgers or recording books. As the media historian Lisa Gitelman argues in her media history of documents, Books like these were for writing, or at least for incremental filling in, filling up. Fillability in some cases suggests a moral economy (diaries and fern and moss albums, for example), and in many others it suggests the cash economy with which nineteenth-century Americans had grown familiar. Filling up evidently helped people locate goods, map transactions, and transfer value, while it also helped them to locate themselves or others within or against the sites, practices, and institutions that helped to structure daily life.2

Taken directly from pre-existing forms of record keeping in institutions, blank ledgers accounted for the principal preservation and finding system in the Smithsonian and later the United States National Museum (USNM). These books were blank forms awaiting inscription. In writing object descriptions into pre-existing fields, the museum preserved the information and, ultimately, the meaning from earlier forms. In keeping with Gitelman’s observations about the preservation of vernacular habits, blank ledgers maintained particular affordances that were shaped by earlier media forms whose epistemic loyalties lay in accountability. Do these books encode particular value systems in and of themselves? What histories do these forms carry with them, as objects became specimens through their encoding? They certainly established the rules for entry into the museum, even though these rules were regularly broken. These ledgers allow for objects to become documents, to become “specimens” as fully accounted for and legible in a specific, pervasive discourse. This chapter, and the one that follows, looks closely at media specificity and form. I want to suggest that, as early curators at the Smithsonian also did, objects without a system of registration had no value in the collection, and that, through entry into the USNM’s documentation system, objects that were considered mundane were transformed into what we think of as “objects of knowledge” – things imbued with meaning.

66

On the Margins

Figure 2  Workroom for Indian ethnology, A&I building, ca. 1890 | Smithsonian Institution Archives, Image NHB-3680

Effectively, what was transformed was not the object itself, but the object as a metonym for entire communities or ideas. Specimens – supported by bureaucratic practices – became valuable precisely because of their records.

Inscription, Encoding, and “Trusted Media” When the Smithsonian distributed its field guides or circulars to military officers and travellers, the USNM already had a rudimentary system of re­ cording to keep track of and document its collection. Items were combined loosely by date and collector, groupings that later became known as accessions. Objects that had been physically sent to the museum by boat, train, or post were sorted in large rooms, sometimes years after arrival. At the USNM, the information about them was recorded in what were variously called “recording books,” “catalogue books,” or “ledger books” – what Candace Greene has called the “trusted medium” of this time period.3 These large bound volumes, which contained spaces to make lists, are akin to many earlier forms of accounting documentation. Figure 2 depicts the

P a p e r S y s t e m s o f C l a s s i f i c at i o n

67

ethnology workroom circa 1890, where staff are “working up” the collection by logging objects in books. Such volumes were common to many early museums, and examples of early catalogue books exist across Europe and North America, but the exact origin of the ledger book as a recording device in museums has yet to be fully traced.4 At least since the late eighteenth century, European museums commonly recorded objects in their collection in lists.5 For example, at the Museum Volkenkunde in Leiden, in the Netherlands, the original catalogue book or inventory list was called the “Inventaris van het Cabinet Rariteiten nagelaten door Mevrouw J.L. van Oldenbarneveld genaamd Tullingh Weduwe van Den Heere J.T. Royer” (Inventory of the collection of curiosities bequeathed by Mrs. J.L. van Oldenbarneveld, named Tulling, Widow of Mr. J.T. Royer).6 The museum also includes books written by people on trading ships. At the Pitt Rivers Museum in Oxford, another early ethnological museum, the inventory was called the “List of Anthropological Objects Transferred from the Ashmolean to the Pitt Rivers, 1886,” and consisted of lists of object names with descriptions and, occasionally, small drawings. Ledger books, as forms, were never fully realized – they were in a constant state of flux. The information that was written in them was modified and changed, and sometimes ignored. Yet their form remained remarkably stable and was outsourced and copied by other institutions. In Candace Greene’s detailed discussion of the ledger books in early American museums, she compares the early fields in the Smithsonian ledgers to those of other institutions, including Harvard’s Peabody Museum, the American Museum of Natural History, and the Field Museum of Natural History. Her analysis details the similarities between these forms and ledgers. A useful comparison can be found in the Field Museum of Natural History in Chicago, where ledger books similar to those at the Smithsonian existed, although they were developed much later than those at the Washington institution, and perhaps were inspired by them. The “Department Inventory” of the Field Museum begins with A1 – “A” standing for “anthropology.” The first volume includes objects bearing the numbers 1–4,000. On the first page, an identifying note tells us that this record is of “South America 1,” with other notes from 1972 saying that the “total number of specimens” is more accurately 4,428. The fields are as follows: When received; Catalog number; Original or accession number; Object; Locality; Number of specimens; Received from; By gift, loan, or purchase; Collected by; When collected; Dimensions or weight; and Remarks. The first entry, from 23 June

68

On the Margins

1904 reads: “A1. Tin Heads of figurines Ref. Cinacontl [sic]; reddish pottery,” with the locality of “San Juan Tcotilnocan [sic], Mexico, Aztec, Nahuatlian [sic],” and the number of specimens listed as five. For the rest of the specimens on the page, ditto marks are used in place of repetitive location names, and the word “pottery” is circled. The items under A1 were collected by F.H. Sellers, and are listed as a gift. Other objects from the same collection are listed with material types – “Wooden Bows” and “hard wood bark covered” – circled and other parts of the entries are also circled or underlined. Other books at the museum have the same headings. Objects are inscribed much later than the date at which they arrived at the museum, and it is likely that the books were used as an inventory of the collection long after the original accessions dates. The object numbered A14301, a “Totemic Carving on bone” from Alaska collected by Ed. E. Ayers, was received in 1894 but not documented in the ledger until twenty years later. Unlike the Smithsonian ledgers – whose fields show change through time – the Field Museum ledgers remain the same, and I surmise that these were all written as an inventory by the same person or group of cataloguers. As Greene observes, these early ledgers in Chicago make use of almost exactly the same field headings as the Smithsonian ledgers, which come much earlier in time.7 Reading them also shows curious engagements with the blank bureaucratic list as a form – for example, markings, drawings, and ditto marks that confuse the otherwise orderly data. In the Field Museum, these ledgers show that material type and object type are privileged, in part because the corresponding card catalogue is organized by object type, rather than serially (like the Smithsonian). The card catalogue is not a replica of the inventory ledgers, but a complementary system with a distinct organization.

Form of Function: Ledger Book Fields through Time At the Smithsonian, the first ledger books were in all likelihood conceived and written by Spencer F. Baird. At the beginning of his career, he had been both a naturalist and an esteemed ornithologist.8 Baird’s tenure at the Smith­ sonian began when he was quite young, and his work involved developing a method for sorting through collections.9 Baird had written some of the first catalogues of natural history specimens including for his personal favourite, the ornithological collections, and in October 1858, he had published “Catalogue of North American Birds, Chiefly in the Museum of the Smith­ sonian Institution.”10 For him, the collection of material objects was akin

P a p e r S y s t e m s o f C l a s s i f i c at i o n

69

to that of natural history materials. He sought to arrange the collection systematically; otherwise he believed objects would be unintelligible and without value. He strove to emulate proper scientific methods, arguing that the collections should be “subject to scientific supervision – can be arranged in proper order – duly classed and cataloged; carefully labeled – so as to be intelligible to all.”11 Baird was appointed the Smithsonian Institution’s secretary in 1872. In that role, he was dedicated to keeping an order to the collections. Indeed, Candace Greene concludes that Baird, along with Frederic Ward Putnam of Harvard’s Peabody Museum, was particularly influential in the use of the ledger system. The fields developed for object entry across many early American institutions were a result of the spreading of documentation ideologies. Baird felt that it was imperative that objects that had not yet been in­ corporated into the documentary record of the museum were never to “be entrusted to inexperienced persons.”12 Inventorying and maintaining a catalogue of the collections were critical, and allowed a range of new scientific research to occur. However, the job of cataloguing and registering every museum object became increasingly time consuming, and not every object was entered as soon as it came into the institution. As discussed in Chap­ter 1, having information directly from the collectors was important, as it would link the objects to “facts,” even if they were not catalogued until years later.13 Recording information about objects in ledger books validated the information supplied in collectors’ lists and field books. Yet reading the ledg­ers at the Smithsonian exposes the tensions between the ideals of data and ob­­ject collection and the practicalities and routine work of record keeping. These volumes reveal the limitations of the application of a scientific, naturalhistorical approach to collecting and recording information about objects. Using circulars to guide the collection of information and objects did not necessarily mean that good, validated information was entered into the initial record-keeping system, or even that it had been collected in the first place. An analysis of the ledger books needs to take into account their information infrastructure – the labels and fields denoted, the space allotments for each piece of information, and the frequency with which these categories were applied and used. At the Smithsonian, the organization and development of the early ledgers were first explicitly mentioned in the 1889 annual report, which states only that the catalogue had been devised entirely by Spencer Baird in 1839 to document his personal ornithological collection, and that it was later adopted

70

On the Margins

by the museum generally.14 Baird developed the systematic practice of record keeping by noting objects in ledger books, what he called the “systematic registration” of the museum objects. The first mention of this process in the museum as a whole was in the annual report for 1857. When objects were received they were: recorded in a general record book ... The different specimens are next labeled and then entered on the record for the class, or particular order, and from this posted in a ledger consisting of separate sheets, one for each species, systematically arranged, and each sheet containing an enumeration of all the specimens of its species, with the localities, sex, date, measurements and other memoranda, making the third time of writing out the name and statistics. In this way not only can information be obtained of the number of species of each class or order, but also of the separate specimens, with the locality and general character of each one.15

I found many of the first draft ledgers and what appear to be order forms for the printing of more books in the archives of the Smithsonian. Early drafts of the “National Museum Register – Marine Invertebrates” include the headings “Family,” “Genus,” and “Species” across the top of the page, followed by columns of subfields. The “Catalogue no.” was different from the “Original no.,” which had been given to the specimen in the field. The category “No. spec[imens]” showed that there could be many parts to a specimen. “Locality” referenced the physical location of the find; “Collector” was the individual who plucked the specimen from the ground (or, in this case, the sea); and “Rec’d from” would have been the individual who physically dropped the specimen off at the institution. “Dry or alc.” referred to the mode of preservation (specimens preserved in alcohol are now referred to as wet collections). The other categories – “Alive or dead,” “Depth, fath[oms],” “Nature of bottom,” “Temp. Fahr[enheit],” “Age and sex,” and “Ovicapsules” – all relate to the specific kind of specimen. These fields are similar to what seems to be a draft zoological ledger from the mid-1880s in the same archives. These kinds of registers are more than the simple catalogues or inventories seen in other museums, as they include a standard set of data to be entered. The first ledger in use at the Smithsonian to catalogue the ethnological collection was simply called the “Museum Catalogue Book,” and it likely no longer exists in the same form that it did in the 1850s. The first ledger

P a p e r S y s t e m s o f C l a s s i f i c at i o n

71

Figure 3  Copy of first pages of original ledgers, anthropology ledger books, volume 16, c. 1859 | Department of Anthropology, Smithsonian Institution, Museum Support Centre

in the Department of Anthropology’s storeroom, labelled Volume 1, has an entry date of 1859 (see Figure 3). Its first entry is an object, simply given the number 1 and labelled “Japan Silk Red,” that was (as the ledger states) given as a present from the emperor of Japan to Commodore Matthew Perry.16 The fields in this book were very similar to those in the earlier zoological and marine invertebrate drafts in the archives: Current number, Original number, Name, Sex, Locality, When collected, Nature of object, Corresponding number of, Measurement, Received from, Collected by, Cost, When entered, Number of specimens, and Remarks. Duplicate copies of each ledger book were made, but it is often difficult to determine when this happened. In the copied version of the first ledger, the word “Locality” is crossed out and “Tribe” is written above it – and both versions contain different information within that category. Smithsonian ledger headings changed through time (see Table 2): changes were often handwritten to reflect the kinds of information and evidence required of ethnographic specimens. There were two copies of many of the

72

On the Margins

ledger volumes made as well, and there are slight differences between them. In the case of the first ledgers (ca. 1859), the original copies had fifteen headings or data entry points. The inclusion of a “sex” column, and its maintenance for roughly thirty years, demonstrates that these fields were not only borrowed from natural history paradigms and information requirements but were direct copies of other ledgers already in use at the time or were modelled after the zoological or marine invertebrate ledgers. In 1867, fifteen fields remained, but “station” (which presumably meant “field station” at the time) was excluded, as well as the field “prepared by.” These terms and methods of description either fell out of use or were consciously excluded to allow for other categories like “Corresponding number” and “Received from.” For example, the field “Prepared by” was perhaps dropped to reflect more common practices in ethnographic collecting whereby specimens were not perceived to need preparations in the same way that other biological or natural history specimens did. This perception is in stark contrast to the description of the preservation of human remains in the circulars discussed in the preceding chapter. The addition of new terminologies permitted the Smithsonian, by then twenty years old, to track not only the “current number” and “original number” but, by the 1890s, the “accession number” as well. This is evidence of a new museum registration process whereby accessions were recorded as part of a consecutive numbering system and perhaps by the formal establishment of museum procedures after the move to the UNSM building in 1881. By 1899, the fields in the ledgers were modified once again and the number of columns was reduced to twelve. The ledgers now contained the following columns: Museum number, Original number, Accession number, Name, Locality, People, How acquired, Measurement, Referred to, When entered, Number of specimens, and Remarks. These ledgers show a reduction in the contextual, site-specific information in some of the earlier field guides, which may have resulted in an unintentional loss of data for the objects catalogued during that time period. They also differ from the initial recording system developed by Spencer Baird, most notably with the dropping of the “Sex” column. The 1899 version of the ledgers represents a simplified, perhaps more anthropologically focused, approach to understanding and documenting collections. “Accession number” appears as a category, and the ambiguous “Corresponding number of ” from the 1867 system has been dropped. Fur­ ther, “Museum number” is used in place of the previous “Current number.” The previous categories of “When collected,” “Nature of object,” “Collected

P a p e r S y s t e m s o f C l a s s i f i c at i o n

73

Table 2

Smithsonian and USNM ledger headings through time 1859

1867

1899

Current number Original number

Current number Original number

Name Sex Locality

Name Sex Locality

Museum no. Original no. Accession no. Name

Station Nature of object When collected

Nature of object When collected Corresponding number of

Measurement Collected by

Measurement Collected by Received from

Prepared by Cost When entered No. of specimens Remarks

Cost When entered No. of specimens Remarks

Locality People How acquired

Referred to Measurement

When entered Number of specimens Remarks

by,” “Received from,” and “Cost” were condensed into the “How acquired” column. By 1899, “When collected” was changed to the column “When entered,” and an important piece of documentary evidence was lost in this transition. The date of collection was written out of the history of the object – a reduction of the object’s origins to the date when it was catalogued in the museum, and a reconfirmation of the notion that these objects are of value and evidence only once they enter the collection and become useful or available for scholarship.17 The way the ledgers changed through time, particularly near the end of the nineteenth century, is relevant in the context of the practice of work at the institution, and is part of the ongoing legacy of these and later records and practices. The ledgers include handwritten modifications to the headings, further indication that the Ethnology Department took existing books from other

74

On the Margins

departments and modified them for its own use. For example, in 1859, the headings in the ledger volumes included Current number; Original num­ ber; Name; Sex; Locality; Station (with “Tribe” written above it); Nature of objects; When collected; Measurement (which is struck through); Collected by; Prepared by; Cost; When entered; Number of specimens; and a small space for remarks. The small, yet significant, change of replacing “Station” with tribal affiliation shows that the individuals responsible for cataloguing the objects questioned some of the categories of natural history collection and worked to define new descriptive apparatuses. This change calls into question the assumed omnipotence of information infrastructures; the practices of “tweaking” indicate how humans effect change in documentation systems. As Bowker and Star have observed in another context, nurses who worked with and “tweaked” the medical classification scheme did so in line with their socio-cultural contexts and understandings.18 One thing that complicates much of the information in the ledgers is that the “field site” or “field station” for early ethnologists was not a geographical location but a community of people. Current museum conceptions of tribal affiliation arose in part from this mode of organization as well. Despite the fact that many objects were given to Smithsonian field collectors by individuals, the ledger books include no space for the naming of these people or the makers of the objects. What is absent from this archival object record? How can we read it, as Ann Stoler suggests, against the grain?19 Early guides urged collectors to pay attention to many details of an object’s origin, but these details are not reflected in the documentary apparatus used to de­fine or label objects within the collection. Marginalia, or handwritten inscriptions that modify the ledgers, are key indicators that show how actors constantly disrupt the discursive stability or form of the ledger through time. Some columns were not used, some were variously used, and there is no strategic approach to naming conventions applied to the columns and fields.

Marginalia: Drawings as Data In 1867, soon after Baird had established the ledgers, he hired Edward Foreman, who became responsible for entering a significant number of ob­ jects into the ledgers.20 Foreman was an officer of the institution and a gen­eral assistant from 1849 to 1885. He had an interest in illustrating the documentation on ethnological subjects, and his cataloguing technique included drawing detailed illustrations of objects in the ledgers (see Figure 4). These

P a p e r S y s t e m s o f C l a s s i f i c at i o n

75

Figure 4  Example of illustrations in ledger volumes by Edward Foreman, ca. 1884 | Anthropology Ledger Books, Department of Anthropology, Smithsonian Institution, Museum Support Center, E88970-E8900

images of objects can be considered one of the first visualizations of the museum objects within the catalogue, a feature that would not appear again until photographs were taken of the collection. It is not surprising that Foreman considered images of objects to be important parts of the documentation process. Even though the ledgers themselves listed specific fields and had material space requirements, these drawings were one of the ways that an individual cataloguer modified conventions of cataloguing and added another layer of object description. Further, Foreman saw these drawings as a kind of insurance against misidentification. In a letter to Baird in 1878, he notes, “I have never, as intimated, in any instance omitted making sketches of objects received, knowing too well that they give the only certain data for making a proper return to owners; the only safeguard against meddlesome and purblind officials.”21 It is unclear why Foreman was concerned with whether or not these objects could be returned to their original owners, but he certainly saw image capturing in the form of sketching as a major supplement to textual documentation. His ledger drawings are reminiscent of, and perhaps directly inspired by, the practices of object drawing in archaeology or the drawings of scientific type specimens found in catalogue books as early as the sixteenth century.22 Stephanie Moser, who has examined the history of the

76

On the Margins

standards and conventions of archaeological object drawing, observes that creating visual representations of objects was “fundamental to the introduction of classificatory procedures in archaeology, enabling recognition of object types.”23 She links the visual representations in archaeological drawings with the organization of museum objects on display to argue that, com­bined, these “groupings” marked the emergence of a science of material culture. Archaeological drawings, like biological drawings of nature, were systematized and emphasized specific aspects of the objects.24 Yet the ledger drawings represented in the Smithsonian’s catalogues do not adhere to those standards; they are more akin to referential marginalia drawings for easy and quick reference to the objects.25 “Marginalia” is both a critical concept and an object of study.26 Critical theories of hypertext, for example, observe in it a Bakhtinian multi-vocality of texts that create a non-linear approach to reading.27 Although commonly referring to electronic texts, hypertext theory has been used to approach archival marginalia in, for example, medieval manuscripts. These marginalia exhibit characteristics similar to the components of hypertext proposed by Landow, such as non-linearity, multi-vocality, and de-centredness. Hyper­ text theory has also been used to argue that knowledges are made manifest not as written texts but within objects, such as wampum belts, thus privileging non-Western and non-written ways of knowing.28 I read some of the Smithsonian’s anthropological ledgers as hypertextual documents. Despite their seemingly rigid and linear organization into tables, the individuals (such as Edward Foreman) who were responsible for inputting information consistently modified and manipulated the forms. Written all over and above the ledgers themselves are decades of notes from collections staff, documentation of object loans, and more. Object drawings in ledger books are also evidence of the attempt or desire to depict the objects and the dependence and authority of visualization as part of the informal practice of ethnology. It is important to emphasize that these documents were created by individuals, who consistently and constantly move outside of the paper form to modify rules when needed. Revisions, although interesting from an institutional historical perspective, are valuable indicators that enable us to understand how research and revision were practised at the institution throughout its history. Handwritten modifications appear regularly in the ledgers, and shorthand was often used to replace full text, owing to the

P a p e r S y s t e m s o f C l a s s i f i c at i o n

77

repetitive nature of the work. Frequently in the physical ledgers, ditto marks stand in for objects’ names, collectors’ names, place of collection, and more, which can lead to repetitive entries and ultimately mistakes in attribution, something particularly relevant to repatriation cases.29 Notations range from brief revisions of locality information to the addition of descriptive notes. In addition, the “remarks” fields often included loan information and other general remarks. Occasionally, these revisions include the name of the individual who wrote them. One notation from July 1928 reads that the red entries were entered by a Miss Marian Powys. Another way in which individuals affected the content of the ledgers was by not filling in certain fields. As noted above, because these ledgers were originally modelled on other catalogues from the institution, not all fields necessarily applied to or were deemed useful for ethnographic material. For example, the ledgers’ “sex” column was rarely filled in. Despite the desire to reproduce the same methods for the study and documentation of all objects, ethnographic objects presented a unique case, with unique documentation requirements, an issue that extends throughout the history of the department.

From Object to Specimen: Classification in Practice – The Case of the Hoonah Repatriation Collection In 1884, Lieutenant Timothy Dixon Bolles collected and packed into two boxes sixty-eight objects while assigned as a Smithsonian field collector in southeastern Alaska. Bolles was not a trained ethnologist, and his primary work was to survey the Alaskan-Canadian border, not to collect cultural objects from the communities he met there. He collected these objects from the village of Hoonah, and his particular interest was collecting from shaman’s graves. As he writes: Owing to the custom among the tribes of this part of the country of bury­ ing with their “Shaman” or “Medicine Men” and Chiefs various articles of purely Indian manufacture, these graves when found are prizes. They are shunned by the natives and their contents suffered to decay and crumble away. I was lucky enough to find two of them lately both very old so that only the wooden and leather articles were left. These I have packed and hold ready to send either by steamer or after I get to San Francisco. I have also a dozen or two masks which came from the same place. These I hold for you if you desire.30

78

On the Margins

Bolles sent the two boxes of objects he had collected directly to Washington, to be delivered to Otis Mason, who was then the curator of the Department of Ethnology in the newly created Division of Anthropology. The boxes included packing slips and lists of the items he collected, and the contents were listed on a packing slip supplied to the department. Lists provided with an 1884 shipment included the following information: Contents of Box 1: Robes Aprons Head ornaments War dance clubs Rattles Clappers All known to be taken from graves over 100 years old.

Contents of Box 2: 11 Masks 1 Rattle Bone carvings 1 Glacial pebbles from Marble Island, Glacier Bay showing deposit of marble in mud 2 garnet rock – particles 3 garnet rock – collecting into seed garnets 4 garnet rock – garnets as formed 5 twin garnets 6 mother and baby (open carefully) Stone pestle.31

The objects would have been slowly unpacked, labelled, and classified by a small team of dedicated assistants working in the ethnology department within the Division of Anthropology at this time. Lieutenant Bolles came to Washington on 31 December 1885 and began work and research on the collection with Otis Mason and his team of assistants. Mason’s curatorial annual report notes that Bolles joined the department to “work up the Eskimo material for the tribes of SE Alaska.”32 In January 1886, Bolles was making a list of the entire collection of the “Eskimo” specimens, and he was present most days working alongside Mason. Bolles returned to the department in varying capacities until 1888, when Mason’s records state that he was assigned elsewhere by the navy.33 Bolles also aided in installing the Eskimo collection in the gallery. However, his primary work was to create a catalogue of the “Eskimo” collections in the department. Bolles’s catalogue, entitled “A Preliminary Catalogue of the Eskimo Col­ lection in the U.S. National Museum, Arranged Geographically and by Uses,” was published in the 1887 USNM annual report. As reflected in its

P a p e r S y s t e m s o f C l a s s i f i c at i o n

79

title, it lists all of the “Eskimo” objects that exist in the museum by use and geographical location.34 Catalogues like these were often arranged geographically, and they enabled scientific researchers or ethnologists to know what was in the collection and provided access to the objects that were not on exhibit. Such catalogues were an important part of the scientific exchange of information for museums at the time. As Bolles says of the catalogue, “Nearly two years have been occupied by the compiler in reaching a result which he trusts will aid the student who may be inclined to an exhaustive comparison between the various villages or the Indian tribes whose boundaries touch the Eskimo territory.”35 Although Bolles was not a trained academic or researcher, it seems that his expertise was considered legitimate because of the length of time he spent living in these territories.36 It is evident that he was working directly with the objects and the ledger books of the collection. He also identified persistent problems with misidentification in his research: “In many cases the location or use of an object was wrongly given by collectors; this was misleading at first, but before long the multiplication of these errors caused serious confusion. These have been placed in their proper locality.”37 To support his research, Bolles would have used scientific sources as references. In his catalogue, he wrote that in order to “properly” name the geographic localities of objects, he relied on a specific “nomenclature” provided by Ivan Petroff in his report on the tenth US census. The records indicate that Bolles personally “handled, located, arranged, marked and labeled” a total of 12,293 objects from eighty-six different places. It seems he was comparing objects with those from other places, such as Mexico and Japan (he examined, for example, two objects from Japan and identified one “labret” from Mexico). These objects either had been previously catalogued as “Eskimo” (and thus were included in the collection by chance), or they were being used by Bolles for direct comparison with the Alaskan objects. Bolles provides a small glossary of object terms, all classified by their “use.” For example, he notes six different types of knives – for carving objects, stabbing, and general cutting, and for use with blubber, fish, and snow. Other object types listed were usually described according to their use as well, such as “boxes for spear points” and “tools to skin seal.” However, when describing ceremonial objects or figurative work, Bolles uses the descriptor for how it was made – for example, “carved fish” and “carved seals.” Curiously, several specific object types that we know were included in the collection seem to be missing from his glossary – for example, dance rattles. Were these simply named

80

On the Margins

some other thing, and therefore their history at this point in time is left mute? Were they included in the category “carved birds”? The example of Bolles carefully going through the collections and classifying them by their “use” shows how research was conducted at the museum during the time period when Otis Mason was in charge of the collection. Researchers at this moment were working explicitly and actively to classify the collections to make them more valuable and useful for research, nearly two decades after Spencer Baird created the original ledgers used at the institution. Bolles presumably relied heavily on these books for information about the specific objects in the collection. He classified the objects he found in the collections based on their recorded information, and to some extent his personal knowledge, which enabled their use as scientific specimens. Counting, labelling, and numbering thus improved their value; and the focus of the Bolles catalogue is the total count of objects by geographic location.38 “How many?” seemed to have been one of the most important questions to answer. Perhaps this was because of the assumption that researchers would then be able to discern the “missing links” in the Smithsonian collection and be able to procure those objects that the institution lacked. Bolles’s catalogue, therefore, not only shows us the origins of a practice of record keeping in museums but also traces how objects that were being used in communities became objects of knowledge or specimens. The prac­ tice of cataloguing created objects of knowledge based on, in many cases, predefined characteristics, as objects were intentionally grouped together based on similarity of use. Researchers used lists of similarity and difference to define and categorize what belonged to certain object categories and In­ digenous communities, relying almost entirely on the information recorded in the ledger books as the basis for these decisions. Although it is impossible to tell what was left out of the collections, or what was not collected, it is possible to look at what came into the museum’s collections. Doing so raises several questions: Did the field guides have an effect on how ethnologists were actually thinking and writing about material culture at the time? Did these guides impose categories that then became standards? What kinds of epistemic understandings and abilities did the documentation afford? Bolles was neither a trained ethnologist nor an Indigenous community member, yet the example of this collection gives us insight into how documentation worked in practice. Recording information in the ledgers allowed it to become authoritative. Yet there are examples in the Hoonah Repatriation Collection where names were applied haphazardly or by mistake. In the

P a p e r S y s t e m s o f C l a s s i f i c at i o n

81

ledger entries that record both of the boxes sent by Bolles, object names denoting “rattles” are variously recorded, occasionally emphasizing the formal qualities of the object and what it depicts, and occasionally recording its method of creation – in this case, “carved.” The following is a full list of the various “object names” applied to this collection and provided in the ledger entries: Carving (wood) " " (mask) Beak for duck (Carving) " Mask Carved Wooden Images Breast Armor Wooden Rattle Wooden Rattles " " – carved " " Clapper Wooden Rattle – carved duck Carved Wooden Rattle Wooden Rattle Carved Models of Masks Wooden Mask " " (models) Wooden Dance Wand War Club

Bone Boat Hook Bone Charms Shaman’s Basket and Charms Leather robe sash (fringed) Kilt Leather Shirt (small) Shawl or Cloak Sash Sashes, of fringed leather Shirt Sashes, of fringed leather Necklace Part of Head Dress (wood) Head Dress (grass) Head Dress (leather band and horns upright stick?) Grass Hat Grass indicators of rank.39

Over thirty different names are attributed to the sixty-eight numbered items in the catalogue. When two objects are seen to have the same “object name,” ditto marks are used to indicate that these are the same kind of object. This led to some objects being catalogued as the “same,” or parts of the same, objects on the later catalogue cards, evidence of a desire to find true ethnological “type specimens.” For example, two objects in the ledger book are both called “shirts” and are given the same catalogue card, indicating that they are considered to be two of the same kind and, effectively, parts of the same object. The terms for “rattle” are among the most obviously inconsistent: there are seven ways that the rattles are described in the ledger books, including “carved wooden rattle” to “wooden carved rattle.”

82

On the Margins

The “rattles” in the Hoonah Repatriation Collection were explicitly used as evidence in other reports, such as Albert Parker Niblack’s 1888 report, “The Coast Indians of Southern Alaska and Northern British Columbia.” Niblack’s report is evidence of how ethnological research attempted to slot objects into categories, with varying degrees of success. It highlights the difficulty researchers encountered with classifying objects into singular categories. For example, Niblack’s report first places the Tlingit rattles under the heading “Rattles, Scrapers and Whistles” as part of “ceremonial paraphernalia.”40 Later, the report describes and classifies them instead as musical instruments, and lists them on a scale from “primitive” to “modern.”41 Niblack identifies two kinds of rattles: snappers and “rattles proper.” The most “primitive” rattle, he says, is composed of two hoops and “strung with the beaks of the puffin.” By contrast, the “usual” form of rattle is a hollow wooden chamber containing a dozen small pebbles. Another object, a “Shamanic Dance Rattle,” is classified in the section on music. Niblack notes that the varieties of such rattles are too great and shows a few “typical” examples in his report. Niblack placed this kind of musical production at the primitive end of the scale of progress, but he did note that the topic deserved “special study” by ethnologists in the future: “It cannot be said that in a musical way, according to our standard, these Indians have made much progress, but the music, such as it is, has the nature of an accompaniment to their dancing, or is at least subordinate to other forms of entertainment.”42 Why were these rattles classified as musical instruments? The modernist functional scheme of finding and attributing meaning to form seems to be an explanation, and aligns more closely with Otis Mason’s original 1875 classificatory work, which placed “rattles” within the category of musical instruments. Museum staff believed that considering these rattles alongside those from other places enabled researchers to create an evolutionary story about the objects. Occasionally, names in Indigenous languages were recorded in the field (as was specified by the guides), and these were included in early ledger documentation. For example, a Hoonah rattle is listed as “wooden rattle” on the catalogue card, but in the ledger book is recorded as “carved rattle, Ceco’q” – the latter term a Tlingit word that museum staff possibly mis­ applied to these rattles.43 This information may have been entered into the ledger much later, and was probably due to a later designation as “Ceco’q,” based on new research, or a report like that of Niblack’s. In this case, the

P a p e r S y s t e m s o f C l a s s i f i c at i o n

83

rattle’s use is inscribed on a label directly on the rattle. It reads: “iii. Rattle used by secret societies Boas.” The Hoonah repatriation report notes that, because of the fact that anthropologist Franz Boas wrote about secret societies in his career, this object label might actually simply act as a citation to this work.44 Yet it may be evidence that this was considered to be a type of rattle that Boas documented or was like those published in Boas’s research. The classification of these “ethnological” objects provides a glance at the knowledge practice of typical early ethnological work. This practice worked to formalize the classifications of these objects as types of the same kind, as ethnographic specimens. As I have argued earlier, in order to classify, it is necessary to distinguish and separate out relevant or “essential” characteristics of objects. For these rattles, it was not that the shamanic aspect was denied outright (as it was noted by Bolles in his packing slips), but that the ledger books and other documentation privileged their form and function, and that these sources were used as the basis for future identifications. Of the sixty-three objects represented by forty-seven catalogue numbers identified in the repatriation report, nine objects are either missing or have been transferred out of the collection owing to specimen exchange. In early documentation practices, defining and classifying objects in the Hoonah Re­patriation Collection as “kinds” or “types” and giving them the same name allowed the Smithsonian to justify their status as duplicate objects that could be exchanged with other institutions. This practice, common in the institution in the late nineteenth and early twentieth centuries, manifested as the practice of marking “duplicate” specimens as multiples of already existing objects and occasionally giving them the same catalogue number. As Catherine Nichols has argued, the “processing” of museum collections made them into scientific specimens, and similar objects were considered copies of others, and allowed for mass redistribution of ethnographic type specimens.45 Timothy Bolles even remarked, in a letter to the Smithsonian, that he feared that the objects he was collecting might be of little value because the institution might already have too many – the presumption here is too many of the same kind.46 As certain objects were considered “duplicates” and recorded as such within the catalogue, they were likely distributed to or permanently exchanged with other museums.47 The case of the Hoonah Repatriation Collection best exemplifies many of the processes and procedures that occurred within cataloguing practice in the late nineteenth century. Tracing this collection shows, in a small way, how the history of documentation can delimit what is possible in current

84

On the Margins

frameworks – in this case, that of repatriation. The collection provides an example of how the fluidity of museum documentation came to maintain durable authority and longevity – in part, because it is the only written re­ cord of the exchange or removal of material culture.

Document Boundaries Spencer Baird had enormous and enduring influence over classification and recording practices at the Smithsonian. Baird eventually worked closely with Otis Mason, and their combined efforts helped to organize and classify the collection. As one contemporary observer characterized their work, “The faculty for organization that characterized Baird with the directive enthusiasm of Mason conspired to forward the study of ethnology in its Smithsonian under the much [sic] favorable auspices.”48 The bound ledger volumes were essentially intake forms. They specified the essential fields of data that were to be recorded in the official museum record, information that was prescribed by the circulars or the idiosyncratic views of collectors. In many cases, objects were not recorded until years after their collection, further divorcing them from any original sources of information. At the time of the first entry in the ethnological catalogue, in 1859, ledger books contained the basic information about the objects that had been collected in the field, a practice common across all departments in the institution. The epistemic commitment of natural history and its associated descriptive apparatus of field recording and visual observation created the natural ethnographic specimen and required that good records be kept. Ledger books provided a kind of bureaucratic power to achieve a system whereby objects could be retrieved and used as transportable objects of knowledge, where before they had been unsorted in storerooms of the museum. Without good documentation, and without proper classification within a system, the objects could not be used as evidence in scientific research or properly verified and authenticated. Perhaps it was envisioned that the creation of a ledger system of documentation would enable future researchers to investigate the collections once the tribes themselves had disappeared.49 Above all else, an analysis of the early guides and the ledgers demonstrates the importance of material documentation to the practice of scientific ethnology itself. Yet, despite these claims for authority, and despite the clear desire to have a scientifically accurate way of collecting objects and information from the field and preserving it in the documentation of the institution, there is

P a p e r S y s t e m s o f C l a s s i f i c at i o n

85

often little standardization or consistency in the way objects were documented and classified in the ledger books. This is evidence of a tension between the stability of the form of the documentation and the actual (in practice) fluidity and malleability of pen or pencil on paper. In their work on scientific objectivity more broadly, Lorraine Daston and Peter Galison argue that the practice of recording information through visualizations (object drawings of type specimens, for example) worked to reconcile two realities: the scientific objective reality and the reality of nature.50 The documentation of objects was a way that museum researchers near the turn of the century could establish museum work as scientific work or scientific ethnology. At certain points in time, characteristics of the objects were pulled into focus and documented rigorously, and these descriptions acted as pieces of evidence in research. Objects from communities became, at different points in the history of the collection of the USNM, the “working objects” of anthropology, both materially and ontologically. Museum objects can therefore be described as the “natural objects” of the work of ethnography throughout the latter part of the nineteenth century. Objects were therefore not simply collected and described but were crafted to become “useful” specimens. Defining desiderata in the field and using ledger books and other recording systems are only part of the rigorous recording practices in the history of anthropology. The way “ethnological specimens” were documented under­ went significant changes near the end of the nineteenth century, as more formalized museum practices took hold in the Smithsonian. By 1899 Otis Mason had become an active figure in the Department of Ethnology, who, throughout his career, sought to make sense out of the “mess” he saw within the museum’s recording system. He is credited with the adoption of the card catalogue system by the museum – a universal ordering system still referred to by staff and researchers who work with the anthropology collections to this day.

86

On the Margins

3

Ordering Devices and Indian Files

Cataloguing Ethnographic Specimens

By the end of the nineteenth century, the world was awash in paper. – Alex Wright, Cataloging the World

In the back of the Anthropology Department in the Museum Support Centre of the National Museum of Natural History (NMNH) sits a small room filled with cabinets, books, loose papers. Atop one of the shelves is a small box, labelled “Indian File,” that contains decades-old cards with groups of tribal names on them, next to lists of numbers and names that show where, in the other massive card catalogue cabinets, one might find the collections relating to a particular community. The aptly named box (Figure 5), although small, purports to arrange a large collection of objects and information by tribal nomenclature. The Indian File was keyed to the department’s card catalogues, which contain almost two million numerically organized cards. The file was a card index file created, at some point in the organization’s history, to sort parts of the Indigenous ethnology collection by arranging them by the names of the Indigenous individuals associated with the objects or collections. Before the creation of this index, because the card catalogue was organized numerically, it was likely very difficult to search for or find the name of an individual creator, maker, or community. The collections at the NMNH have, for the most part, always been listed and organized by accession and catalogue number, which often reflect a

Figure 5  The “Indian File” in the Department of Anthropology of the National Museum of Natural History | Photo courtesy of Diana Marsh

particular collector or expedition. This larger card catalogue is effectively organized by the order in which boxes of objects were opened in the Smith­ sonian Institution’s attics or warehouses many decades earlier. The only indices that organize the collections information in a different way are the Tribal-Locality File, the Musical Instrument File, and the Indian File. Such indexes clearly enabled researchers to gain access to the collection in different ways, although it is difficult to tell how long the Indian File was used. It brings to light the importance of understanding how organizing cards in a file system comes to represent ways of thinking about collections. In many ways, this card catalogue system as a whole is indeed an Indian File – a bureaucratic practice and mechanism for sorting, classifying, and organizing people, their belongings, and their ancestors. Through the magic of bureaucracy, objects – given away by, or taken or stolen from, peoples from all over the world – stand in for people, ideas, practices, communities. The magic extends, too, to the mechanism of the card catalogue, created to organize and control information. In this chapter, I show how the Smith­ sonian’s Department of Anthropology began to organize its files in the late nineteenth and early twentieth centuries. From the 1870s to the early 1900s, the ledger book recording system developed by Spencer Baird was still used as the principal source of information about objects in the Department of Anthropology. In 1884, Otis Mason became curator of the Division of Ethnology, and, over the course of his career, he significantly altered the process by which objects were recorded and documented. The adoption of a card catalogue system in the Department of Anthropology happened during his tenure, and it is likely that Mason

88

Ordering Devices and Indian Files

was primarily responsible for this innovation, as he was an outspoken advocate of such “universal ordering devices.” The years after 1875, with the creation of the United States National Museum (USNM), proved to be an important time of change for the Smithsonian, and the collections, staff, and exhibits spaces were moved into two new buildings in 1881 and again in 1911. In the thirty-year period between 1880 and 1910, a number of pub­lic exhibitions and world fairs had a significant impact on the development of the collections, more circulars were published and distributed, and different sets of requirements and new, more specific, foci were introduced.1

Card Catalogues as “Paper Machines” Card catalogues have held a special place for historians of libraries, technologies, and information. As indicated in a now famous passage by Foucault, the card catalogue has been both a mainstay of research and an elusive historical subject: “The appearance of the card-index,” he writes, “and the constitution of the human sciences are another invention that historians have taken little note of.”2 One of the most significant developments for museums was the adoption of the card catalogue, which is both a technology and a methodology for sorting and compiling all of the information about objects in an easy-to-read, easy-to-access, movable, indexable system. As col­ lections increased in size, they were seen as more valuable for research, but they soon required a more robust system of information management. The history of card catalogues is long, arguably beginning with the use of paper slips organized in cabinets at various points in history. As early as the sixteenth century in Europe, card indexes were developed to manage large repositories of information and books. Markus Krajewski describes the Swiss polymath Konrad Gessner as the “father of the modern bibliography.”3 Gessner published the first volume of his Bibliotheca Universalis in 1545. This record of the works of three thousand authors describes the texts in a way that is very similar to modern bibliographies: authors, dates, formats, titles, and places of publications.4 He devised a method where mobile paper scraps could be affixed to a special type of book with a guiding thread – the paper slips could be moved, replaced, filled in – but stored in boxes to remain safe.5 This tension – between durability and flexibility – is one at the heart of the information system of the card index. Card indexes originated in Europe but were widely adopted in libraries and sold commercially as early as the mid-1800s in North America. Soon,

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

89

they were adopted as ubiquitous office-management resources. The origins of the library card catalogue are credited to work at the Harvard College Library, which, in 1812, hired William Croswell to develop a cataloguing system.6 Following Croswell’s work in organizing the library’s collection, Harvard hired a man named Ezra Abbot to make the card catalogue more accessible by adding headings and labels to the drawers of cards, by increasing standardization of the card size, and, importantly, by making the catalogue publicly available in 1862.7 Smithsonian librarian Charles Coffin Jewett had heard about Croswell’s work at Harvard, and he pursued the idea for a similar catalogue for the Smithsonian. In 1853, Jewett announced his cardbased catalogue at the first international librarian conference in New York, indicating his desire to collect all United States bibliographic information in one card index. Jewett’s ambition, if not his proposed method, would in­spire Melvil Dewey.8 By the 1870s, Dewey would make cataloguing systems and technologies commercially available, with his company, the Library Bureau.9 By the late nineteenth and early twentieth century, cataloguing developments in library science, including those applied in the library at the Smith­ sonian, were influencing museum cataloguing practices. Indeed, it is possible that the card catalogue system used in the Smithsonian libraries in the mid-nineteenth century directly inspired USNM departments to consider applying that technology to organize information about its collections. But the Smithsonian was not alone in its desire to develop a card catalogue to hold and retrieve information about its collections, and, by shortly after the turn of the century, there were many other examples of card catalogues in museums.10 The ways that the information was catalogued on cards, into accession files, and on the ledger books differed – sometimes fundamentally – among museums. The card catalogue systems that were forming in American museums soon became a topic of interest for German anthropologist A.B. Meyer, who noted that these card catalogues were based on Dewey’s standards, which were common in libraries.11 By that time, the scholarly consensus was that museum workers should have library training to help them deal with the complexities of cataloguing.12 One of the key components of the technology of the card catalogue was its ability to hold information on “separate, uniform, and mobile carriers [that] can be further arranged and processed according to strict systems of order.”13 In the library, the card catalogue provided the abstract representation of, and controlled access to, texts. As I explore in this chapter, the USNM used such catalogues to represent information about cultural heritage

90

Ordering Devices and Indian Files

and, by proxy, many of the communities whose belongings were taken into the museum. The card catalogue is an organizing device, and I suggest that this method and its material affordances worked to stabilize the ideas, forms, and nomenclatures that had been variously and non-systematically applied in ethnological work until this time. Previous paper tools like the field guides and ledger books had been used to keep track of and tally the collection and information about it. The ethnological card catalogue, as envisaged by Otis Mason, was to be used to cure the symptom of disorganization, which he saw as a detriment to the scientific work of the museum.

“A Severe Reckoning”: Organizing Chaos The 1880s brought an increase in staff and collections to the USNM, encouraging curators to begin to systematize the “unwritten laws of the establishment” in order to facilitate the “routine work” of accessioning and collecting.14 A new “plan of organization” was implemented, and the offices and workrooms were completely rearranged. Further, the Office of

Figure 6  Storage area in the United States National Museum, early 1880s | Smithsonian Institution Archives, Image MNH-2603

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

91

the Registrar was established, and it became responsible for sending packages, managing the storerooms, and recording accessions (see Figure 6). The authority to affix numbers to objects, and to label and describe them, still rested with the curators, but there were no enforced standards of procedure. Although the administrators at the museum wanted a routinized system of classification and documentation, this was difficult to achieve on a practical level. The best practices of cataloguing and associated documentation became a crucial topic for the Division of Anthropology at the USNM in the late 1800s. Evidence of the development of such practices can be found in reports such as Spencer Baird’s annual report for the year 1880, where he lists the complete journal of an object from the “field” to the storage location in the museum. This journey is telling. First, explorers or field collectors would “announce” that they had sent certain collections from specific localities. Then, likely a cataloguing assistant, or perhaps a curator, made an entry in an “announcement book,” whose headings included the date of announcement, the name and address of the sender, the point from where the collec­ tion was sent, the general nature of the collection, the agent who forwarded it, and the method of transportation.15 When the object or collection arrived at the museum, it was recorded in the department preparation room (the preparation room for Indian ethnology can be seen in Figure 2 in Chapter 2) and in an accession register, where a donation number was attached to it. If any additional information accompanied the collection, it was given to the curator of the collection, who would later sort the objects and review any accompanying information. Otis Mason was an important figure in the museum at this time, and his influence on the collections extended beyond his published circulars that we saw in Chapter 1. He published several scholarly monographs on the sub­ ject of the USNM’s collections, in which he often looked at and compared object types across locations and time periods.16 However, his influence (or his potential influence) was reduced after Boasian anthropology became more widely accepted in the field.17 Throughout his career, Mason was dedicated to establishing standards and methods of classification, and he was primarily concerned with organizing objects into their functional units. He was also committed to the strict organization of the museum and its collections. Organizing the collection based on a typological method of studying material culture was part of Mason’s method that allowed him, and others, to come to conclusions about an object’s origins and associated

92

Ordering Devices and Indian Files

peoples primarily through comparison. Mason envisioned the new technology of the card catalogue as an index of this system – one that could grant a kind of scientific accuracy to the evolving science of ethnology. Making specimens eloquent, he argued, involved proper collection, description, and arrangement in what can be called an information system.18 The conditions of possibility that characterized this system of record keeping came to form the foundation of the information infrastructure in the Department of Anthropology – and they influence that infrastructure even until this day. Mason was influenced by other work and perspectives at other museums and was motivated by a quest to bring more rigour and higher standards to the developing discipline of anthropology at the Smithsonian. When he began his work to organize the collection in the late nineteenth century, he relied on the arrangement of the physical specimens themselves to encourage research and comparative study. For him, the proper typological arrangement would help explain the development of objects, tools, and practices. As the museum historian Tony Bennett has argued, this typological approach was foundational to the way Mason conceived the arrangement and display of objects at the turn of the century and is a hallmark of what has been called the “classical” natural-historical episteme.19 Mason’s work as a curator at the museum from 1884 until his death in 1908 has been the subject of much scholarly and historical reflection.20 Mason believed that it would be possible to understand the history of human development by placing all examples of a type of object in a progression, and his tenure at the USNM reflected this belief. He wrote with passion about his desire for proper order in the collections, and this was reflected in his writings about his intention to organize the museum in a way that would allow for scientific research to be conducted in an orderly fashion. The ability to locate the objects in the collection was his prime concern, and it inspired his work to encourage the Department of Anthropology to adopt a card catalogue or card index system. When he achieved a paid position with the USNM in 1884, he was appalled at the state of affairs of the record-keeping system: In handling the collection I am more and more convinced that a vast amount of material has been lost or ruined and I would like to bring the whole business to a severe reckoning. By this I mean that I would like to get another lady clerk for a few months to help Miss Latham to organize a thorough system of cards for a catalog and get a correct history of the objects.21

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

93

When Mason moved into his office in the new USNM building and assumed his curatorship, he felt that the ethnographic collection “could hardly be in worse confusion.”22 Under his instructions, objects were unpacked from boxes that had been in the museum for decades, and were then sorted, labelled, classified, and catalogued. Mason began his curatorship by organizing and cataloguing the ethnographic objects that were under his care. Because of his desire to adopt a card catalogue index system and standardize the documentation of ethnographic objects, he played an important role in developing new museum administrative practices. Changes in collections care were motivated by his desire to create an organized and efficient system out of a chaotic one. Mason provides a description of the cataloguing practice in a letter to Baird in 1883.23 As he states, the registrar was responsible for writing the information on the descriptive cards. Collections were given an accession number, and all of the manuscripts related to the accession were filed together by the registrar. This individual was writing this information with direct access to manuscripts prepared for the accession, although it is unclear if the registrar had access to the objects. Although there is no direct reference to what these “manuscripts” could be, it is possible that Mason was describing research documents prepared by field anthropologists, or the field catalogues and notes of the collectors of the objects. If the manuscripts were misleading, or outright incorrect, the registrars may not have been aware of such problems; their task was simply to copy the information onto the ledger and the descriptive cards, which in turn served as a source of definitive knowledge on the object itself.24 Often what were considered “ethnological” objects were sent to the museum accompanied by many other kinds of specimens, like geological and biological ones. When the specimens arrived at the museum, the collection was categorized and sent to separate departments. The ethnological specimens were sent to the Department of Ethnology and given a museum number, which was attached or written directly on the specimen, and then an entry was made in the “catalogue” (the ledger books). After that, templates for “descriptive cards” were printed for each individual specimen, and the information from the ledger book was written thereon. Securing the data was important, of course, but Mason was well aware that, without proper instruction, those who were responsible for collecting the data might make egregious errors. In Mason’s mind, such errors would decrease the value of the collection: as he noted, “precious old relics in museums have lost a large percentage of their value, because their origin is

94

Ordering Devices and Indian Files

unknown.”25 The “science” of natural history was still used as the basis upon which the “new science of cultural history” was modelled. Mason maintained that the natural historical sciences were always careful to note the associated “data” for their specimens, particularly the location, collector, and donor.26 For both the museum and the individual curator, the “best results” in this case were publication and exhibition of the objects. Mason was dedicated to maintaining the scientific accuracy of the collections and their associated data. This is in part why he was adamant that the field-collecting guides published by the Smithsonian be used by collectors – he saw the latter as the only “trained experts.” The systematic collecting advocated by the field guides was a hallmark of scientific collecting for Mason and, along with good documentation, gave the entire collection a value and purpose. Mason’s personal diary from 1884, when he was made curator in the department, reveals much more detailed information about the practice of cataloguing. His notes reveal that the workforce was a small group that was engaged primarily in sorting through old boxes of objects and caring for newly arrived material, much like staff in the other departments in the museum.27 The Smithsonian annual report for the year 1884 contains a special section entitled “Progress in Classification and Arrangement,” in which there are clear parameters set out for the work of the curators in their “private apartments,” where collections were “received, unpacked, classified, identified, and cataloged.”28 The classification of objects was primarily done by the curator’s assistants, who were making “customary entries” into the ledger books and, in Mason’s words, “preparing a card catalog in which all information that can be obtained in reference to them, is embodied.”29 The work done by the staff, curators, and researchers who entered the collection was seen as indispensable intellectual work by the Smithsonian and other early ethnologists, and the labels prepared by researchers were considered the ultimate source of knowledge “upon which all future authentications depend.”30

Good Order, Good Data Just as the anthropological collections were growing under Mason’s care, the collections of the libraries at the Smithsonian had also grown in number and had become an essential part of conducting anthropological research. At the same time, there were increased discussions concerning the classification of scientific material in the Smithsonian libraries. Mason wanted to

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

95

properly classify the anthropological collection to ensure that “good” scientific research was possible. Principally, he was concerned with creating a comparative, typological approach to cataloguing ethnological objects. In an 1884 letter to Spencer Baird, Mason expressed his hope of facilitating research for individuals and curators through the comparative study of specimens. He advocated for the use of what he considered to be a proper scientific method for the arrangement and description of specimens to support comparative analysis, which, in his mind, required “well-authenticated” artifacts. He saw this process of arrangement and description as essential not just for study, but also for broader education, part of the institutional focus on scientific research and public knowledge production.31 Mason set out to create an index that identified the function, material, and location of every object, particularly for the American ethnological materials. Mason referred to the card catalogue as a “card encyclopedia.” He saw it as more than a list of contextual information about the objects, as normally would have been established by the collector. For him, these cards should contain verified and well-documented information about the objects and their place within the larger classificatory schemes. Mason intended the physical arrangement in drawers to match the classificatory concepts he developed to ease the burden on the researcher or visitor who needed to see (and engage with) the reference specimens in the museum. He argued that the advantage of this method was that his assistants could easily find objects, and that “specimens badly defined or without labels have an easy explanation by means of their nearest neighbours.”32 The nature of the encyclopedic card catalogue, which made information easy to retrieve, allowed vacancies or missing pieces to become visible, and enabled the museum to identify objects to be collected. As Mason highlights in his reference to the circumpolar collections, “All the objects in the Eskimo collection being placed in their appropriate boxes, the vacancies will be at once apparent and proclaim either that the people of that area do not use this device or that the national museum has not the good fortune to possess one example.”33 Identifying these “obvious” and “natural” gaps would allow the museum (in particular Mason) to request specific types of objects through field guides and circulars. For Mason, part of achieving a proper scientific method was the classification of ethnographic objects by their function through this typological “encyclopaedic method”:

96

Ordering Devices and Indian Files

Whenever a collection of special importance is received from the same locality, the objects are first carefully classified so as to bring all things together that are alike or that have the same use. These are then entered as formerly, a separate number being given to each piece that is sufficiently distinct to receive it, but those that are alike or that form a set receive the same number. The card catalog is used also with this material for the purpose of gathering information. In the register things go by number, in the card catalog they are arranged by topics and classes of things, so that all information upon each subject will be found together as in an encyclopaedia. This encyclopaedic method has proved of incalculable value in the correspondence of the museum when difficult questions are proposed for immediate reply.34

More importantly, attributing the proper classification to an object would therefore give it value. In his annual report from 1885–86, Mason elaborates: “The effort is ... made to give each object all the scientific value which it possesses from the very start and to render it accessible at once for the exhibition series, the study or type series, and the exchange series.”35 Mason desired to create a fully “classified” report of all of the ethnological materials, designed according to ethnological categories. According to Mason, these were “function, tribe, geographical distribution, degree of elaboration, material, and classes of investigation.”36 This meant rearranging the objects on exhibition and in storage to “facilitate ready reference and comparative study.”37 The object storage was known as the “study series,” where more serious study of the objects was done. Objects that arrived from the same location or community were distributed among the collections. Objects that “went together” were considered simply to “be alike” or to have the same use or function.38 The way that objects were made into specimens of value was through comparison. The late nineteenth century marked a move away from understanding singular, “monstrous,” or unique objects as being particularly valuable, which had been the norm in previous museum practice.39 In George Brown Goode’s “Recent Advances in the Museum Method,” published in the 1893 Smithsonian annual report, he argued that these single specimens were in fact much less useful than a specimen series: “Single or unrelated specimens, though valuable or interesting, are in themselves of little moment in comparison with series of much less precious objects which unite to teach some lesson to student or visitor.”40

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

97

The degree to which objects were similar would be the most important organizing principle of the storage drawers for study, exhibition, and the arrangement of the card catalogue. Similitude or difference would make it possible not only to fill missing gaps or vacancies in the collection, but also to make up for the fact that, most of the time, there was simply not enough information accompanying any given specimen. Known data from similar objects was believed to be transferable to objects lacking data. As Mason continues, One of the greatest difficulties which a curator has to encounter is that which arises from false location and insufficient data ... This is one of the chief hindrances to a purely ethnological specimen, since it is often begging the whole question to assign a specimen to a certain tribe. On the other hand, no harm can possibly come from putting things that are alike in the same case or receptacle.41

This association by similarity of form provides a key insight into early cataloguing at the museum. In one respect, without the basic documentation listed in the ledgers and on the catalogue cards, Mason felt that objects had little use or value; on the other, he saw that their physical and informational arrangement could lead to discoveries about their use. Effectively, their physical arrangements could lead to conclusions about their history, and these conclusions were printed and made available to the public through labels on the displays. However, Mason still felt that, without direct access to the original “authenticated” information from the collector, establishing the value of the objects for scientific research would be extremely difficult. From his perspective, situated in evolutionary anthropology, it was considered good method to simply put things that were “alike” together, and this was seen as a legitimate avenue to developing museum knowledge. As noted above, objects that appeared to be the same received the same catalogue number. Mason’s card catalogue proved useful in this context because, as he notes, in the catalogue the objects were “arranged by topics and classes of things, so that all information upon each subject will be found together as in an encyclopaedia.”42 Mason was influenced by the work of European museums, such as the Pitt Rivers Museum and the British Museum, which he visited on a trip to Europe during his tenure as curator,43 and the Museum of Ethnology in Leipzig. He envisioned a kind of card-based system, something derived

98

Ordering Devices and Indian Files

from the principles of museum classification established by Gustav Klemm in Leipzig.44 Mason also noted the importance of the cataloguing system in the Department of Ethnology in the British Museum: “The method of cataloguing is worthy of imitation. Each specimen is accompanied by a large sized catalog card, which bears in addition to ample description and identification, a good drawing.”45 This trip has been well documented,46 and his travel to many museums and conferences during this sojourn proved influential for his development of the encyclopedic catalogue, and for his vision of the museum as a whole.47 Yet there is no documentation in the Smithsonian archives or the annual reports that explicitly states that he modelled the department’s card catalogue after the one at the Pitt Rivers Museum in Oxford or the British Museum. The general developments in library card cataloguing during this period were likely more influential for the museum’s practice. The physical arrangement of cards in space was necessary for researchers, but standards for how to name the specimens were also needed. In his 1884 curatorial report, Mason outlines specifically how this nomenclature should be applied to specimens, but also outlines the challenges: Among the ends which the Curator of the Department of Ethnology is very anxious to accomplish and for which he begs the cooperation of his scientific brethren, no one seems to be more important than nomenclature of the different classes of objects with which he has to deal. On the one hand an utter disregard of nomenclature is so vicious that no argument needs to be urged against it. On the other hand there is a danger of overloading the subject with too many difficult names lending rather to confusion than to perspicuity.48

This passage highlights the tension between and the performance of Mason’s organization and the durability of his organizational system. Nam­ ing strategies were not regulated, but were performed continually and in different ways as each object was recorded, described, and labelled. On the records that were documented during Mason’s time, there are, for example, various spellings of tribal names and of object names. Objects identified as “Mask” are categorized as “Wooden Mask” or “Mask, wooden.” Consistency was the goal, but the practical nature of the work required fluidity. In his diary, Mason notes his daily work practices and some strategies he had employed in cataloguing, which began by arranging things simply by “their name.” For example, on his first day as curator, he “spent the day in sorting

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

99

material putting into the same drawer ... things which have the same name, whether this arises from the structure or the function.”49 The ethnology collections were large, and arranged not geographically but by type. Mason believed that collections were puzzles that, when aligned accordingly, could provide answers. He later continues: “The most impressive fact is this, that we go on collecting things [that] are alike and by and by a specimen turns up in which the things are in their structural relation. This solves the riddle at once.”50 Mason’s typological method of classifying objects had an effect on how entire sets of objects were named and classified. For example, in Mason’s Ethnological Directions, “Rattle” is classified under the subheading “Instru­ ments for beating,” which appears as part of the overall category “Music”: (11.) MUSIC. A. Instruments for beating. Rattles. Clappers. Bells, sounding-bars, &c. Drums. Tambourines. Marimba. Others not mentioned.51

Here, Mason’s approach was to put objects in relation to their use as musical instruments, not, for example, their associated method of manufacture or their visual characteristics. Grouping all “rattles” under the classification “instruments for beating” denies differences between different kinds of rattles and other mechanisms. Despite this problem, terminology needed to remain the same within this system in order to facilitate access to the objects through the records. The arrangement of the card catalogue allowed researchers to search by heading or object type. A researcher could flip through the cards and immediately obtain the name and some descriptive information, which would have been impossible under previous methods of registration. Grouping by similarity of form was a basic component of what has been termed a Renaissance episteme – that information could be gleaned from relationships of similitude.52 Objects were grouped according to hidden relationships, and thus it became possible to know both the visible and the

100

Ordering Devices and Indian Files

invisible. As Hooper-Greenhill argues, this knowing was the entire goal of the museum itself as a system of knowledge, and museum display was the tool that enabled this interpretation.53 Classification plays a fundamental role in what Foucault terms the classical episteme.54 Similitude, or the same­ ness of things, was contrasted with difference, and large classificatory schemes were developed based on the interplay between the two. As HooperGreenhill argues, to know was to discriminate, a practice clearly articulated by Mason in his strategies of object documentation.55 Mason’s reverence for classificatory knowledge organization, and his desire to order the collection, was based in earlier scientific practices that were already out of popular academic use when he began classifying the collection. Indeed, the classical episteme was not a neat temporal distinction but rather a way of knowing that persisted in some spheres longer than others. It is as if Mason was dedicated to making the study of human beings and their material culture as scientific as possible. Doing so meant, for him, that arrangement and classification generated true or accurate knowledge. To return to the example of the rattle, for Mason, its form is its capacity to produce music, and this was the only categorization that mattered: it would have been nonsensical to him to classify it in any other way. Yet the narrowness of this classification denies its actual use as a ceremonial healing object and is, essentially, a denial of the affective relationships objects and belongings had with communities, a denial that is then replicated in the catalogue. Catherine Nichols’s recent examination of Otis Mason’s influence on the distribution of duplicate specimens at the museum argues that Mason was generally in favour of classificatory concepts being applied to objects and that he sought to arrange things in the manner that a natural historian developing a taxonomy would. As Nichols has shown in her study of pottery classifications from this time, Mason employed four concepts when examining collections: he looked at place, form, function, and material.56 Although he drew up and imagined large, classificatory typologies of objects, her research demonstrates that there was no one way in which Mason determined which objects were to be exchanged as duplicates. Mason’s focus on typologies of civilization’s progress was physically embodied not just in the display of artifacts but in the storage of them as well. The objects that lacked appropriate labelling and contextual information then gained value by being placed in assumed natural arrangements. The development of the card catalogue then allowed these objects to be indexed and searchable, without there being a need to view their arrangements in space, and it allowed researchers and

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

101

curators the ability to record information about the objects that did not perish once the display was taken down.

Formalizing the Ethnographic Card Catalogue: 1890 and After With the help of an assistant, Mason worked to develop a new card catalogue. These cards, available in large cabinet drawers and sorted by catalogue numbers, were to represent each object in the ethnological collection in a one-to-one ratio. Objects were still entered into the ledgers first, and were given an accession number and then a catalogue number. Accompanying the cards, at the front of each “accession” (a group of objects that were entered into the catalogue at the same time), was a card that read “Key to the Collection.” These cards provided a general description of the entire group of cards (and hence, objects) associated with that accession. By late 1890, the card cataloguing system was being used more regularly, and there was a collaborative effort by Mason and his assistants to retrieve all of the information from collectors and transfer it onto descriptive cards. In 1890 and 1891, a printer produced copies of blank cards for the catalogue, providing the form through which Mason would achieve his goal of having a more permanent system of organization in place. A full catalogue of objects in the Department of Ethnology had never been attempted. A complete card catalogue would, as Mason saw it, enable future curators to put their hands upon all that was known concerning each specimen. It would also give additional value to all objects sent out for exchange.57 By 1895, Mason had formalized his ideas concerning the encyclopedic catalogue. Each card was to include five different data point entries: the catalogue number, the name of the object, the structure and function of the object, the place and tribe from which it was collected, and the name of the collector. These data points worked to make the ethnographic object legible within the wider discipline of the museum and within anthropological inquiry. Yet, in 1896, Mason was still thoroughly frustrated with the state of the collections as a research tool and wanted the museum to cease collecting everything brought in by donors. Instead, he wanted the museum to determine the “missing links” or gaps within the collection already at the USNM and fill these in through specific collecting missions. Mason attempted to pursue this end by first achieving a “correct understanding of what constitutes valuable ethnological material.”58 He felt the USNM was falling behind other museums and argued that “no pains should be spared to procure this

102

Ordering Devices and Indian Files

material” to fill the gaps in his typological series. This goal no doubt motivated him in co-drafting a second version of his Ethnological Directions circular with William Henry Holmes in 1902. By 1898, the card catalogue system was in full use by the Department of Anthropology staff, and it had become part of normalized practice. The ledgers or catalogue books were kept by a “recording clerk,” and the work of recording and classifying the collections in several separate divisions was now compiled by a single “expert assistant.” Individual curators or keepers were to make individual card catalogues for their own research as well, organized however they wished. By 1900, cataloguing was given its own section in the curator’s annual reports, and card catalogues were adopted as the primary way of organizing almost any kind of information relating to the collections in the department. Mason’s description gives insight into the daily routine of a curator at the turn of the century: In writing this report the head curator has assembled on his desk the following data: the department’s books recording accessions, permanent and temporary, for the year; the card catalog of accessions; card catalog of collections for the year from all the divisions and sections; (each tied up in bundles, each representing an accession or necessary subdivision of an accession, with detailed data, statistical, historical and individual); the card catalog of storage; the card catalog of exchanges; the card catalog of articles poisoned or otherwise treated; and a card catalog of articles made for the museum by preparatory or other persons employed by the museum.59

By 1902, the ledgers and cataloguing system of Mason’s encyclopedic catalogue had fully developed and were purportedly “kept up to date.”60 Yet not everyone uncritically embraced Mason’s system. The executive curator at the time, J.W. True, wrote to William Henry Holmes, the head of the Department of Anthropology, because he believed it had weaknesses that needed to be remedied. In a 1900 letter True argues that a total revision to the record-keeping system should be completed, and he raised the possibility of ignoring everything that was written before 1900, owing to data inconsistencies and numbering problems.61 Curators encountered the durability and stability of decades-old mistakes and, despite his frustration with them, enabled their longevity. But, if he had destroyed this legacy data and started anew, there would have been no complete record of objects anywhere, and he ultimately saw this as an even larger organizational problem.

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

103

At the turn of the century, a growing and influential scholarly community was dedicated to the study of museums. The standardization and bureaucratization of practices such as cataloguing and registration became common across many institutions, but the development of standard museum practice was relatively new. The professionalization of the museum discipline in the United States had a great influence on the concern for establishing guide­ lines and more rigorous approaches to accessions and documentation. Im­ port­antly, the American Association of Museums was founded in 1905 by Smithsonian staff; in Britain, the Museums Association began to publish the Museums Journal in 1901. These groups established networks of communication about issues related to practice. In the 1902 issue of the Museums Journal, the Smithsonian’s card catalogue, and the catalogue at the Field Museum in Chicago, received international praise, as a result of which they may have served as models for other collections.62 Nonetheless, after 1903, the Smithsonian’s annual reports make little reference to the actual process of cataloguing.63 In 1904, there was a shift in collections care as a result of the museum’s acquiring the collections of the Army Medical Museum. These were primarily human remains, and, as a consequence, the Department of Anthropology added a Division of Physical Anthropology. This change in organization and growth in both personnel and the collections made it imperative to update and expand the exhibitions as well as the storage for the study series.64 As a result, cataloguing roles within the department were consolidated, and the position of head recording clerk was created. By 1904, the procedures associated with the care of collections, such as preservation and cataloguing, had become routine. At the same time, the systematic arrangement of these collections, particularly in anthropology, was still important, and some of the process of labelling is described in the annual report for that year. Mason’s curator’s report for 1905 includes a detailed description of recent cataloguing practices. He defines the term “accession” for the first time in a detailed way as “a collection of specimens, or, may be, a single specimen, placed in the museum by gift, loan, transfer from some other branch of the government, or deposited by the Smithsonian or other body.”65 He specifically notes various types of information that are required in order to make “specimens eloquent.” In addition to “exhibition halls ... and picture galleries” that a contemporary might associate with a museum, he highlights the importance of “information, card catalogues, labels, reference lists.”66 Here, card catalogues and labels are positioned as, essentially, parts of the

104

Ordering Devices and Indian Files

object itself – and as “sources of knowledge.” The card catalogue was an essential part of collections care. Mason is forceful in his conviction, stating that “everyone who has the interests of the science of our species at heart sufficiently to save the material witness, will confer the greatest benefit by being sure of the data accompanying each specimen.”67 Listing the objects on catalogue cards was the last step in the process of registration. As is noted in the 1905 Smithsonian annual report, after the objects have been “studied and classified they are entered on catalog cards, [and] arranged in a systematic order so as to constitute a descriptive list of the objects in each class, the basis of future research.”68 One of the last full descriptions of the practice of registration and cataloguing appears in the 1907–8 curator’s annual report, which states that “card cataloguing must be carefully done and data secured for labels and monographs.”69 The card catalogues allowed researchers to search for and find information about objects; they would then publish their findings in the Smithsonian’s scien­ tific journals, which the field considered formal, authenticated documen­ tation about the objects. By 1908, the year of Mason’s death, the catalogue card system was in active use, although Mason’s curator’s report for that year spelled out a few precautions about both the classification and care of the collections.70 This report, published in the Smithsonian Institution Annual Reports, includes a section on the “care and classification of the collections,”71 but mention of the actual process and complications of cataloguing, or even reference to the catalogue cards, is absent. The intellectual discussion and focus of collections care soon faded, in part owing to Mason’s death. The focus of the institution turned to the lack of storage space and the desperate need for a new building. Further, the emphasis on the Department of Anthropology was nearly dropped entirely and was replaced with full and detailed reports on the (then) extremely active Bureau of American Ethnol­ ogy. By 1912, the entire USNM collection had been moved into the new building, a process accompanied by intensive labelling and classification.72

“Deadwork”: Routine in the Twentieth Century As the Smithsonian changed in structure and size, so too did work roles in the institution. Curators were no longer solely responsible for all of the documentation or research, and new kinds of museum workers were brought in to help sort and record the collections. These individuals were considered to be non–subject specialists who were doing a simple and mundane task.

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

105

For example, when handwritten catalogue cards were revised and designed specifically for the typewriter, clerk-typists were hired to complete this work. Although they were directly copying the information dictated by the curators, this work was far more than just note taking and typing. Nonetheless, these roles were likely filled by relatively inexperienced workers. For example, the Works Progress Administration (WPA), the Depression-era program that provided unemployed people with jobs in public works, furnished 144 assistants to the Smithsonian in 1939, and 126 in 1940, to assist the museum staff with “miscellaneous activities.”73 This effort helped the museum enormously in continuing its work, particularly in the cataloguing, numbering, and study of the collections themselves. By 1952 the work of the typist was to copy labels on a labelling machine, type data for files, rearrange files and prepare indices, and work with accessions to type the catalogue cards.74 Documentation and record-keeping tasks are often assumed to be mundane and to require no expertise. These tasks often fell to women, and the invisible nature of their jobs has meant that the contribution of women’s work in the museum is often unacknowledged. Women working as clerkstenographers and typists had jobs that related to documentation in libraries and museums. Richard Urban, in his review of museum and library training programs, shows that, in the mid-twentieth century, many of the participants in the American Association of Museums’ training programs were women.75 The role of women as typists during this time has been acknowledged elsewhere,76 and, while it is not the subject of this book, I would note here that, as cataloguing became a routine work practice, it was delegated to those who were seen to be “non-experts” – that is, clerk-typists performing a routine and mundane daily task. A woman named Sarah Latham served as Otis Mason’s personal secretary. In that role, she likely did much of the cataloguing work during his tenure, and her handwriting is scattered throughout the ledgers and cards. Although only ever referred to by her last – what is likely her father’s – name (even in the announcement of her retirement, she is listed as Miss S.E. Latham), she was a key part of the working documentation staff in the department.77 Latham was likely responsible for unpacking boxes and for the initial description of objects, and she worked closely with researchers at the institution. Her role was never described as more than a clerk or copyist, yet she put a lifetime of work into the collection. Her handwriting and labour are present in all of the “good documentation” that was required by male curators and directors.

106

Ordering Devices and Indian Files

The Smithsonian used other kinds of “non-specialists” as well, especially interns. A 1959 letter contains a request for two summer interns to “aid in checking the objects against the catalogue cards and correcting records as well as adding new object numbers.”78 For other types of work, “experts” were commissioned. There was a distinction between the intellectual or scientific work of classification and arrangement, and the routine practice of inscribing numbers and organizing the catalogue cards. For example, a 1959 application notes that a group of northern Native American objects needed to be rearranged and classified, and this was to be completed by someone “above the level of a clerk or ordinary aide.”79 Between 1908 and the early 1920s, information in the archives on the practice of cataloguing in the museum is sparse, perhaps in part because collecting was stalled in the wake of the First World War and the ensuing economic depression. It may also have been because, as Otis Mason had hoped, the practice had become so routine “it scarcely merits repetition.”80 There is some evidence of how collections care was handled, and how specimens were made useful to the department, in the curator’s annual reports archived in the National Anthropological Archives. For example, in 1920, Walter Hough, who had been appointed the acting head of the Department of Anthropology, stressed the routine but necessary work of preparing and cataloguing the collections: “The visible Museum must be kept to the highest point of perfection and the work this necessitates is constant and exacting. In the background is the tremendous routine of occu­ pational activities which the geologist [James Peter] Lesley called ‘deadwork’ and which must be completed before specimens are brought to public view.”81 The larger Smithsonian Institution was actively exchanging objects between schools and other institutions, and the pressures on the curators to publish and to work on exhibits limited the time they had to work on the catalogue. Considered, as Hough puts it, a “minor task,” or in Lesley’s words, “deadwork,” cataloguing was to be done by “preparators,” and Hough wanted to hire more staff members to take up these duties.82 Although curators considered this work to be routine, they also acknowledged that it was necessary to the functioning of the department and were aware of its ability to contribute knowledge to the discipline. The preservation of specimens took a central role in these years. In 1921, Hough notes that the curators were working to perfect their “museum method” and that the preservation of specimens required a large amount of work on the part of staff. This is evidence of a growing epistemic shift in the routinization of museum work.

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

107

Figure 7  Walter Hough, in his office, 1908 | Smithsonian Institution Archives, Image SIA2012-6450

Figure 8  Walter Hough and staff examine textiles from a new ethnological collection, ca. 1908 | Smithsonian Institution Archives, Image SIA2102-6448

108

Ordering Devices and Indian Files

Figure 9  George D. McCoy in the Catalogue and Recording Office, ca. 1920 | Smithsonian Institution Archives, Image SIA2102-6564

In the mid-1920s, the emerging bureaucratic practices of the institution had significant effects on how objects were catalogued and cared for. The cataloguing and the description of objects were seen as part of non-intellectual work or discovery – a routinized practice designed to care for the information of the objects as property of the institution. Ideas about cataloguing as “deadwork” and a “minor task” reinforce a move away from curatorial oversight of collections documentation. Figures 7, 8, and 9 show evidence of a bureaucratic culture heavily influenced by the card catalogue and paper­ work. Figure 8 is especially poignant, with the portraits of Native Americans looking down from the shelves above the large catalogue cabinet onto those responsible for classifying the collection. Researchers not only viewed the collections on exhibit and in storage but actively conducted research on groups of objects as well. The department continued to take great care in classifying and arranging these objects. Staff created lists that located all of the material in the study collections, and, as Hough noted, this made it possible “to locate, with a minimum of effort, any individual specimen not on exhibition.”83 However, it was still impossible to search the collection by collector, and this proved to be a significant

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

109

problem, particularly with respect to the archaeological collections. This limitation was perhaps due to the financial problems of the institution. The annual report for 1923 notes that financial resources had dwindled, and it was “impossible for the museum ... to increase the scope of its exhibits and of their usefulness to the public.”84 In spite of this, the museum continued to be interested in filling the gaps in existing collections, and once again the emphasis was on the classification and arrangement of specimens to increase their value. Hough acknowledges that much of the data on the existing col­lection had been lost, that “official” records were missing, and that the accessions were largely disorganized. Although it is unclear if staff were successful, the work of remedying inconsistencies finished, and the department turned its attention to creating an index of collectors so that all of the material from one collector could be located in the study series. Although indexes helped researchers to find objects and their associated information in the collection, documentation was what made these objects specimens of science. Within this mode of knowledge, objects that were not collected “scientifically” were of little use and, as Hough comments, were “falling so far beneath the requirements of modern science.”85 What kinds of scientific evidence were material collections thought to provide at this time? Primarily, they were considered to have practical value – that is, they contributed to knowledge “by increasing the scientific outlook of the world, by performing an educational duty, by preserving evidences of the progress of man in the arts, and by practical supplications to the material welfare of society.”86 They could not perform this duty without proper labelling and without a position in the card catalogue. Hough expresses more dissatisfaction with the progress of the catalogue in 1925, and he notes that the indexed list of collectors was yet to be finished: “There remains in this connection much to be done in identifying wrongly numbered specimens, completing cards improperly or inadequately prepared, and in furthering the considerable undertaking mentioned in our last annual report, namely, assemblage of an indexed list to collectors.”87 By 1928, Hough felt that the division’s records were finally in a workable and serviceable order. However, he describes a growing, if not main, concern of the department: due to undocumented exchange or de-accession, there were still speci­ mens that had no documentation or records, and there were catalogue entries in cases where there were no specimens. In addition, “scraps of paper bearing catalogue numbers have been lost, thus leaving valuable objects without any current means of identification.”88

110

Ordering Devices and Indian Files

By the 1930s, a significant backlog in the cataloguing of the collections had arisen again; the work around this time was directed specifically at renumbering older specimens. As detailed in a memorandum from 1936, a museum aide by the name of Mrs. Ogle was instructed to conduct this task, and it appeared that “she could continue with this project indefinitely.”89 A work order from 1935 shows that the scientific aides were to engage in the duties of checking the specimens against the catalogues to ensure that num­ bers and labels remained consistent, that classification was correct, and that objects were “placed” correctly as well.90 The work of correcting backlogged information was done piecemeal, reflecting the current research desires of the curators and specific projects or exhibitions. Much of the “data checking” involved rectifying entries by checking the accession papers, going back to the original collections, and going to the “ethnologic” literature and to each curator’s individual “ethnographical knowledge.”91 The tribal index and key catalogue cards were continually revised as well, and the corrections done in the ledger and the catalogue card were to be made across all indices, with objects being renumbered if necessary. Revising work was now considered to be part of the routine job of collections management.

Cataloguing after 1930 The next significant changes to the established practice of recording ob­ ject information appeared after the death of Walter Hough in 1934. A 1935 memorandum on changes in accessions described, in detail, the process of cataloguing in the Division of Ethnology. Staff were to remove objects from packing cases and clean them, and then, as soon as was possible, pre­pare a list of the collection to accompany the accession file. The head curator’s office then assigned individual catalogue numbers to each object, based on the information within the accession file and list of specimens. Then the staff stored objects that were fragile or of “intrinsic value,” and they classified, “poisoned” (i.e., treated with insecticide), and cleaned the collection as a whole again.92 At this point, staff measured and described each specimen. With respect to the ethnology catalogue, the goal was to be able to make a “brief ethnological characterization of the specimen, not merely identification from other more or less similar material.”93 As the methods of early anthropology had changed to rely less on material culture collecting and more on direct observation and participation, it was no longer considered “good” research to categorize and thus typify the objects

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

111

in museum collections. The catalogue reflected this particular epistemological shift. Part of the reworking of the cataloguing system in the mid-1930s included switching from the handwritten card index developed by Mason and his contemporaries to typewritten cards. The earlier handwritten cards were deciphered and then copied, a sizeable task. The curator or researcher dictated the description to a clerk-typist, who entered information directly onto the catalogue cards in “Elite” typeface.94 The cards themselves were not incorporated into the “regular catalogue file” until the close of the fiscal year, when “the final place of exhibit or storage is entered on each card in pencil.”95 The curators maintained their individual “card files” of accessions, one arranged by the name of the person (donor) or the institution and another arranged by accession number.96 The card catalogue allowed for many different entry points to the information and, therefore, the objects. Indices were arranged in what were considered to be the most useful for research, which, for anthropologists at this time, was by donor. Accessions – that is (usually) the physical groups or boxed objects that first entered the collection – were still the primary informational organization of the material culture data. The Department of Anthropology instructed its constituent divisions to change the routine of accessioning: each division was provided with its own individual catalogue book, and, as each new collection arrived, the curator of the receiving division would create an accession. However, owing to con­fusion, and because different methods of cataloguing were in use by different curators, by 1936 it was suggested that the divisions’ catalogue books be removed and that only the catalogue card system be kept.97 The curator’s report from that year notes that one particular catalogue, a triballand geographical index of specimens, had been completed by this time.98 This type of index reflects what Mason originally envisioned in his work at the turn of the century: its reorganization of objects by affiliated culture and location, rather than just by donor, was made possible by the mechanism of the card catalogue. Eventually, additional indices of this type acted as guides to the full card catalogue of object records, which were still organized by accession. Notable differences are evident in the records of the divisions of the De­ partment of Anthropology. Archaeological records were grouped numerically, with a cross-reference to collector; the physical anthropological cards were grouped both by number and by geographical area; and ethnological

112

Ordering Devices and Indian Files

Figure 10  Frank M. Setzler prepares archaeological collections for study, ca. 1930 | Smithsonian Institution Archive, Image SIA2012-6456

specimens were grouped alphabetically and cross-referenced to geograph­ ical areas and accession numbers (see Figure 10). For all divisions, before objects were classified, a memorandum of accession was written, likely by curatorial assistants or typists, and the catalogue numbers were assigned afterwards. The description of the objects was then dictated for typing on the catalogue card.99 The discipline of anthropology as a whole had become more specialized, and the requirements for the recording of information about objects had changed. The work of recataloguing and reclassifying the collection continued into the late 1930s. According to the reports of head curator in the De­ partment of Anthropology, Frank Setzler, a new index of tribes and sub­ jects, in loose-leaf folders, was drafted.100 The demand for documentation increased as ethnological research became more specialized. Despite poor records and incomplete information, the reports note that it was the duty of the ethnographers in the museum to understand simply by viewing the object what type of specimen it was and to supply an “exact identification of the tribal and geographical source of an undocumented specimen.”101As

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

113

Setzler notes in his annual report. “Recourse to this knowledge must be resorted to by the staff practically every day of the year.”102 It is clear that multiple, if not conflicting, ways of organizing collections information existed in the Division of Ethnology, as there appear to be shifts in how the index to the catalogue functioned. For example, by 1942, the organization of the catalogue was revised to become a “numerical-specimen” index, owing to increasing fears that collections might be destroyed during air raids.103 As Setzler observed, The division’s catalog always has been arranged geographically and tribally. But not numerical. At the present time we are faced with the possibility that the only available numerical catalog, that in the Head Curator’s office, will be removed from the building as a precaution in case of an air raid. Thus, it has become necessary to change the division’s catalog to a numerical system and to develop a new geographical and tribal index. At the close of the fiscal year Mrs. Murphy had completed the portion of this index pertaining to the states and had arranged the relevant cards numerically.104

The Smithsonian considered the collections information stored in the catalogue to be of such significance that it was to be protected for fear of destruction during the Second World War. This numerical card catalogue was to be moved from the head curator’s office to an off-site storage area with a set of collections that were to be stored in a repository west of the Appalachian Mountains for the duration of the war. Staff members were therefore tasked with creating a complete and second geographical and tribal index, and changing the existing catalogue to a numerical system. It is un­ clear exactly what Setzler meant by “numerical” system; however, the card catalogue today is organized by museum number or catalogue number, which provides good evidence that it is, at the very least, similar to the “numerical” catalogue discussed by the curator. This process of numerically organizing the catalogue caused problems and brought to light some existing inconsistencies in the practice of cataloguing, which are described by Setzler in his annual report for 1942/43: The extensive revisions of the divisions catalog, begun last year, was carried to completion by Mrs. NB Murphy, clerk-stenographer, during the early fall. The divisional catalog cards were checked against the catalog books in the

114

Ordering Devices and Indian Files

head curator’s office, and thus numerous instances were discovered where either the cards had not been entered in the books or were themselves missing. The importance of this work has become more evident since the evacuation of the catalog books.105

Staff members were actively correcting records as well, and the backlogs of revisions and cataloguing could be significant. For example, Setzler notes that “extensive additional documentation of the Cherokee Indian materials collected by James Mooney a half century ago and now included in several museum exhibits has been added to the catalogue cards from Mooney’s original manuscript notes describing the specimens.”106 These backlogs and corrections are evidence of the department’s best efforts to increase standardization and the indexability of the catalogue, and to define and record “core” or “essential” data about objects. Yet, despite these intentions, the constant reorganization and reworking of the cataloguing system in fact proves that the material affordances of the existing card catalogue were not successful in aiding the museum to organize the information about its collections. Retrospectively, we can see the iterative and developmental process of how the cataloguing system changed, highlighting the extremely local, site-specific nature of how documentation works in practice. In the 1950s, the routine of accessioning and cataloguing remained similar to earlier instantiations: objects were still classified, listed in the ledger books, and recorded on catalogue index cards. This process is noted in the archives: “Catalog cards grouped alphabetically according to collector. Cross reference to geographical areas, also to accession number. Upon receipt of specimens, Memorandum of Accession is immediately started before specimens are classified. After catalog numbers have been assigned, information and description of artifact is dictated directly for typing on the card.”107 Further, the catalogue number continued to be printed on the specimen itself. In 1953, the catalogue cards and the associated “History of Collection Cards” were edited. The catalogue cards were to mirror the information available in the ledger books, and the headings on the cards from this era reflect the changes to the headings in the ledgers. The editing process involved a retracing of documentation practices in order to obtain the earlier information on the ledgers. The curator’s annual reports from the 1950s record that cataloguing as a practice had begun to catch up to the rate of accessions, and information was

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

115

written on cards during the same year that the accessions were received.108 There is mention of a Native American tribe coming to the museum to identify photographs, and the information given by the Elders was considered to add “value and usefulness” to the collection.109 Instances such as these were few and far between, though, and the regular practice of authenticating catalogue records was done by the check of data provided by the donors and other “ethnological reports” from the areas where the collections had originated.

Persistent Mistakes: Revising the Typewritten Card Catalogue At certain points in the process of documenting and registering objects, if the information went missing or was invalidated, objects could lose their value as scientific specimens. Information was lost, at least partly, because of the continuous backlog of documentation caused by great influxes of objects, lack of staff, and volume of work. The objects that entered the collection, whether through the Bureau of American Ethnology, through government surveys, or from individual donors, were often accessioned and catalogued long after their arrival in the museum. Such delays caused significant problems. First, the work to “catch up” to the backlog was always great; throughout its history, staff in the Department of Anthropology have felt this pressure. They have consistently needed more staff and volunteers to assist with the workload. Second, delays resulted in a paucity of information about the objects themselves and, in many cases, the entering of incorrect information by those responsible for the accessioning of the objects. A common practice was to give objects that came from the same location and collector the same number.110 With this numbering method, objects often went missing, and information was attributed to others incorrectly. These mistakes persist in the records. Different systems of record keeping and the development of multiple files or indices over the life of the department are indicative of the difficulty of the practice of cataloguing the collection. Information was made durable once inscribed in the ledgers or on the catalogue cards, but the constant iterative development of the process resulted in fluidity. In some cases, this rendered the information useless or completely incorrect. Such problems are explicitly discussed in a letter to Frank Setzler, in which Saul Riesenberg requests the hiring of a clerk-typist to aid in the work of cataloguing in 1957:

116

Ordering Devices and Indian Files

The Division of Ethnology is currently engaged on a revision of its localitytribal file. The present file, except for cards made in the last two or three years, is virtually valueless. The divisions by tribes and places are inconsistent, some areas being organized by culture area, with tribal subdivisions thereunder, others by continent with tribes arranged alphabetically, still others combining both systems, so that one must look in several places to find anything, and may find the same tribe under more than one heading; and some ex­ pectable headings are not to be found. More-over, there is a lack of crossreferencing. The material on any one card is by accession, which is useless for almost all purposes. On each card, under the accession number, is given the full range of catalogue numbers in which the specimens from the particular tribe fall, this range being that of the whole accession, so that often, to find the dozen or so items attributed to that tribe on the card, one must go through several thousand catalogue cards. On the tribal card the entry usually says simply “Ethnologica” or a few items are listed (without reference to catalogue number) followed by “etc.” This makes it impossible to turn to a particular tribe and find the catalogue numbers of all specimens of one type.111

This passage is indicative of the increasing frustration with the system of object documentation in the card catalogue, particularly as a finding aid for the collection. Riesenberg notes the impracticality of having the history cards and “key cards” contain information separate from the individual object cards, as information rarely was transcribed to all relevant categories of cards, and thus the information in the catalogues could be incomplete and un­helpful. As Riesenberg continues, The key cards, which frequently lack information which needs to be filled in from other sources, are more of a hindrance than a help. They project above the level of the catalog cards, but show accession numbers of the cards be­ hind them instead of catalogue numbers, which impedes search, particularly when there are large numbers of them with only a few catalog cards behind them ... Eventually, when an inventory of specimens is made, the superfluous cards must be weeded out. Any letter answered, information prepared, or research done on the basis of information on the catalogue cards, lacks assurance that the specimens are actually in our possession.112

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

117

Figure 11  Draft, typewritten catalogue card | National Anthropological Archives, series 2, Alpha Subject, 1828–1963, box A, folder: Accession Routine

These remarks illustrate two things: first, that staff did not consider the various catalogue files and indices to be the best approach to cataloguing; and, second, that the catalogue cards were central to the production of reports or information about the objects, despite their incoherence. Material documentation needed to be constantly revised and honed. The catalogues were revised throughout the 1950s and 1960s, and typewritten cards were adopted ubiquitously. Drafts of the catalogue cards show the proposed changes to the headings required to properly catalogue the objects. During this period, the Division of Ethnology’s ledger books contained the headings: Museum no., Accession no., Original no., Name, People, Locality, How acquired, Measurement, Referred to, When entered, No. of specimens, and Remarks. It appears that most of these categories are roughly replicated on the catalogue cards (Figure 11). The cards in the catalogue today are the result of the revisions during the time the cards were typewritten in the mid-1950s. As the revision process changed from the early twentieth century to the 1960s, varying forms of the cards were developed, including three that are most common in the department’s card catalogue today. The cards contain three numbers: the catalogue number, the accession number, the original number, and a field

118

Ordering Devices and Indian Files

Figure 12  Catalogue card, 1965 | Department of Anthropology Card Catalog, Type E73781, Department of Anthropology, Smithsonian Institution, Museum Support Center

for “marks.” Other categories included on the cards are Name (most often the name of the object), People, Locality, Collector, Acquired, Date, Placed, Size, and Remarks; the division that the cards belong to is printed at the bottom right-hand side of the card. Also, as noted on a draft card (Figure 11), this card was to be of “heavier stock” than earlier cards. The draft for a card from 1965 (Figure 12) included the category “Neg[ative] no.” in lieu of the “Marks” category, a reflection of the fact that taking photographs of the objects became standard practice in the department after the 1950s. Yet another type of card, also dated to 1965, includes many of the same fields, with an additional “No. of specimens” field (see Figure 14). These card types may have simply been used as new accessions came into the museum, or they may have been used during the slow process of copying and typing the information from earlier handwritten cards. Throughout the 1950s, increasingly detailed statistics on the department’s backlog were kept to document the rate of accessioning and cataloguing. However, if the cards were not typed up properly, the accession was not considered “finished,” and in 1954 a full-time clerk-typist was hired to take on this work. It is important to note that the work of cataloguing was mainly done by a single clerk-typist – the short staffing a consistent source

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

119

of frustration for the department – and that the work of transferring the handwritten cards to the typed ones was considered a task that could easily be done by someone without specialized judgment or knowledge.113 The process of transferring old catalogue cards to newer typewritten cards required research undertaken by other staff. The more detailed work and the “systematic examination” of old collections was done by an assistant curator, Robert Elder, who was responsible for improving the old records and data based on the objects, while the physical examination of the specimens (going back into storage to actually look at the objects) was done by a museum aide named Clarence Bender. At the same time that catalogue cards were being revamped and updated, two types of index card catalogues, which helped organize the collection, were also being reworked. These “History of Collection” and “Key to Col­ lection” cards (Figures 13 and 14) were possibly based on models developed much earlier by Otis Mason. Both of these cards were placed in the file at the beginning of an accession, and the individual cards for all the objects in that group sat behind them. There are many types and different formats of these cards, but all have several commonalities. Other History of Col­ lection cards include space for the range of catalogue numbers to which the card applies, the accession number, the name of the donor and the address, and the date the object was received at the institution. They include the ledger book number that corresponds to the accession, with the date it was entered in the book, and a space to record who entered the information in the collection. The cards also include a large blank space where information could be typed in at a later date. This space was left to accom­modate future research on the provenance of the collection. The Key to Collection cards served a different purpose. The intention was for the cards to provide needed contextual information for the registration of the objects that had all come into the museum as a single accession. Although revisions to the cards were conducted over time, they were never seen to be a perfect system. Because earlier information was copied directly from the ledgers or dictated to a clerk-typist, mistakes were often made, and much had to be revised prior to the computerization of records. During the history of the museum, there were several other card catalogues, which organized information in different ways. Notable examples are the index of tribal names or “Locality-Tribal File,” and the Musical Instruments File. Additionally, each curator kept their own index of their

120

Ordering Devices and Indian Files

Figure 13  Draft History of Collection card | National Anthropological Archives, series 2, Alpha Subject, 1828–1963, box A, folder: Accession Routine

Figure 14  Draft Key to Collection card | National Anthropological Archives, series 2, Alpha Subject, 1828–1963, box A, folder: Accession Routine

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

121

division’s collection within a card catalogue and was therefore perhaps able to organize individual collections in different intellectual or informational modes. The musical instruments card catalogue, which began as an entirely separate project, is an example of an informational scheme that focused on one topic and was based on the functional attributes of the objects. This card index was developed by F.H. Hawley in 1904. As early as 1889, “musical instruments” received their own separate classification in the Smithsonian, as distinct from other kinds of objects desired for acquisition.114 This catalogue represented the way of organizing the collection at that time, as musical instruments were first assigned to their own department within the museum in 1894–95. That department later became part of the Department of Arts and Industries, and then was subsumed into the Department of Anthro­ pology.115 In working with the collection, Hawley created a descriptive catalogue of its contents as well as a card index of the instruments.116 The card index notably contains more robust information than what exists on the history cards, the catalogue cards, or the key indices – in some cases, the information is new or entirely different. These index cards were retyped, and the original handwritten cards thrown out in 1976.117 The Locality-Tribal File, or tribal index file, still exists in a box in the Department of Anthropology today. Although its origin is unknown, evidence and institutional memory suggest that it was introduced during Otis Mason’s tenure at the museum, when the creation of multiple yet distinct subject card catalogues was in process. This file would make it possible to search for objects as they pertained to individual Indigenous communities. Each file drawer is classified by location: for example, “Ethnology – North America – North of Mexico,” and is further divided with cards separating general geographic areas, communities, and then, if the information was available, the specific village. As the tribal index was in the process of being revised, and as Riesenberg had noted, the organization of the material by accession was “useless for almost all purposes.”118 The revision of these cards would take an extended amount of time – nearly two years. Much of the cataloguing work from the 1940s through the 1950s was done in the context of increasing concerns about the status of the storage and preservation of the collections – objects were clearly falling into disrepair. By the 1970s, the museum decided that its storage areas were sim­­ply insufficient, and a new facility was planned. Ultimately, almost the entire

122

Ordering Devices and Indian Files

anthropological collection was relocated to Suitland, Maryland, where it continues to be stored today.

The Museum’s Universal Ordering Device Throughout this chapter, I have detailed the shifts in collections documentation from the period between 1875 and the mid-twentieth century. I have argued that a slow, yet steady epistemological shift occurred over this period during which the Department of Anthropology engaged in a concerted effort to find and document material culture in North America through the adoption of the universal ordering device of Otis Mason’s card catalogue. Mason was a key figure in the history of Smithsonian collections documentation, and his emphasis on the application of classifications and typologies is evident in the way he studied material culture. Not only did he develop complex lists of object types suitable for collection, but he also wrote and distributed guides to field collecting. His long lists described the types of knowledge considered suitable to collect about the material culture brought to the institution. At the same time, his adoption of the card catalogue was one of the first attempts at creating a comprehensive, indexable search system for ethnological collections in North America. Paying attention to the material affordances of a historical system of organization frames the card catalogue as an indexable knowledge machine, or “paper machine,”119 that laid the foundations for future descriptive possibilities at the NMNH. Otis Mason argued that making “specimens eloquent” involved their proper collection, description, and arrangement in an information system. The conditions of possibility that characterized his system of record keeping came to form the foundation of the information infrastructure in the Department of Anthropology. The practices established by Mason and his contemporaries also endure in the department, because information inscribed on catalogue and descriptive cards was transcribed directly into the computing system. Despite attempts to curate and order objects before they entered the collection, through the establishment of field guides and circulars, objects were not fully realized as specimens until they were documented, classified, and registered in the museum. These practices and technologies all helped to craft the notion of the “ethnographic specimen” as an object of evidence in ethnological inquiry, and as reducible to the object’s function, made visible on small index cards. The process of cataloguing was necessary to organizing the specimens – ethnographic

C ata l o g u i n g E t h n o g r a p h i c S p e c i m e n s

123

objects were essentially useless to the researcher if they could not be found within the collection, and good documentation was essential to ensuring their use as evidence. In other words, without their associated contextual information, they could not be used as comparative objects of knowledge. Two key ideas can be extracted from the history of the routine of cataloguing practice in early ethnology. First, the end of the nineteenth century produced a system of value for ethnographic objects whereby documentation was a requirement of object-worthiness, or of value as scientific specimens. Objects became specimens once their required documentation was attributed to them, and this became ingrained and routinized in practice throughout the twentieth century. Second, there existed a tension between consistency, standardization, and repetition alongside constant change and negotiation. Early on, Mason and his contemporaries simply could not decide on a correct nomenclature for the collection, nor on a single classificatory scheme. The card catalogue was crafted to allow for different entry points to the “data” of the object collection. In this way, knowledge was at the fingertips of any qualified researcher. After 1930, the methods scarcely changed, and the provenance fields established earlier became part of the material form of the changing card catalogue. The information about the collection became supremely important: it was something to be protected, cared for and watched carefully. Mapping this history shows the durability of the form of cataloguing, as both a practice and a technology, and it calls our attention to the iterative, generative, and performative qualities of knowledge work in museums. And, of course, such practices had an impact on the future of the card catalogue as well: the machine-entry punch card inventory and the precursor to the modern database.

124

Ordering Devices and Indian Files

4

Pragmatic Classification

The Routine Work of Description after 1950

The artifacts also have advantages over written records of behavior and belief in being concrete, objective, difficult to distort, and little subject to personal or ethnocentric bias ... Anthropological collecting should be properly defined as the acquisition of well-documented artifacts. – William Sturtevant, Guide to Field Collecting of Ethnographic Specimens

Computerization came to many offices and organizations in the late 1960s and early 1970s, changing routine bureaucratic practices dating from a century earlier. For museums, especially the Smithsonian, early computerized proto-databases made it possible to store and retrieve data in fundamentally different ways from what was possible with typewritten records. Early computing also raised two significant challenges for the Smithsonian: the standardization of terminologies and object names, and the efforts to organize and “clean up” historical data. After decades of inconsistent naming strategies composed on paper-based recording systems (like the ledgers and the catalogue cards), the plethora of distinct names for similar objects was confusing and unmanageable. To ensure that a computerized documentation system would function properly, staff needed to draft standards to guide the naming process. Staff created a variety of lists, authority files (files that specify standards of practice and naming), and ad hoc naming conventions

to ease the transition to computer-based collections management. As routines changed and contemporary collections-management practices began to take shape, different issues presented themselves. Data from earlier typed records and handwritten sources often migrated directly into the computerized systems, and became what is known as “legacy data.” Working with historical data raised practical, technical, and epistemological issues. This pragmatism centred on the technological changes enabled in the 1960s by the adoption of the first computer system developed to con­ tain and retrieve information. What came with the possibility or promise of computing in museums was also the knowledge that data collected and recorded perhaps a hundred years earlier would not “fit” into the specific system requirements of punch card databases or new inventories. Unlike in the late nineteenth century, object classifications were not chosen because staff believed in a hierarchical scheme of material culture. Rather, they were chosen because of their functionality or created to solve the technical issues caused by computing and early databases. However, just as in times past, the choice about what terms to use and how to properly describe objects is not neutral, and it tells us something about the way that historic practices remain durable even when specific technologies change. Taken together, practices and tools can hold their shape and form while circulating through space and time. The effect is that ideas, objectivities, and epistemologies are consistent despite changing attitudes and new locations.1 In their form, card catalogues were information-organization devices, and these devices formed the basis of organization for the computerized database. Taking a closer look at how terms that had historically been attributed to objects were managed and standardized, and how naming strategies for cultural heritage were developed, lets us glimpse the socio-historical construction of museum data. Using an analysis of the computerization of the catalogue cards to look deeply at how legacy data became attached to object records in the collection brings a kind of document durability or a colonial presence into view. After the death of Otis Mason in 1908, his card catalogue was used with increasing regularity, and there was actually little revision to the process of cataloguing in the Department of Anthropology up to the advent of computerized databases. However, cataloguing as an information-management procedure became more formalized and routinized, and thus deeply embedded in the bureaucracy of the museum. The focus of anthropological research shifted toward participant observation, while the study of material culture as the object of inquiry became associated largely with archaeological work.

126

P r a g m at i c C l a s s i f i c at i o n

Yet, eventually, the study of objects came back to the fore in anthropology, and so too did the importance of their documentation. This epistemological shift is best exemplified in the curatorial practice and writings of William Sturtevant, an ethnologist who was appointed curator of the Department of Anthropology in the mid-1960s. His impact on museum procedures and the formalization of the study of material culture was deeply felt in the department and in the field more broadly. Tracing the origins of his approach as well as his influence on the development of anthropological cataloguing situates the computerization of “specimen” data within broader anthropological concerns with material culture.

William Sturtevant and the Union Catalogue In 1956, Sturtevant, a trained ethnologist from Yale with a specific interest in linguistics, took up a position with the Bureau of American Ethnology (BAE) at the National Museum of Natural History (NMNH), where he conducted a number of research and writing projects, including extended fieldwork and collecting. Sturtevant is often credited with returning the focus in the anthropological disciplines to material culture studies, something he was a proponent of for his entire life.2 He was interested in understanding “native” points of view and felt, unlike Otis Mason, that categorizing alternative world views and objects in hierarchical and comparative schemes failed to understand or represent them. This difference in opinion and method of working with collections was not accidental. Sturtevant’s practice was located in an entirely different disciplinary and ideological moment. He saw simplification – for example, the simplification of form and function within the typological method – as fundamentally misrepresentative. As William Merrill noted in a biography of the anthropologist, Sturtevant argued that it was a gross over­simplification to assume that general categories like kinship and religion were the “same” across all societies. As Sturtevant claimed, “It has long been evident that a major weakness in anthropology is the underdeveloped condition of ethnographic method. Typologies and generalizations abound, but their descriptive foundations are insecure.”3 In the same article, he posits that the study of material culture should enhance the accuracy of the study of “native states of mind” because objects are distinct “things” that are “readily observable.” For him, cultural practices and forms were unclassifiable and defied easy typological descriptions, but objects were concrete and could be ordered. He continues:

The Routine Work of Description After 1950

127

In material culture the objects classified are concrete and easily examined and usually readily observable in many examples during the time available for normal field work – in contrast to diseases, deities, etc. ... With cultural artifacts the corpus is smaller and the significant features are largely produced by the classifiers and hence should be more distinctive and more readily recognizable; also the ethnographer can here subject at least some of the features to controlled variation in order to test informants’ reactions to their significance.4

The “classifiers” in this case are the individuals who are living within the cultural milieu to which the objects are attributed; the participants in the system of classification belonging to the culture of origin are “actors.” Sturtevant believed that the project of “ethnoscience” (that is, the methods for the analysis of systems of classification and the practice of defining types of ontological categories like types of species) was to point to and manipulate concrete objects that would enable the ethnographer to get at questions of categories and contrasts, instead of solely relying on “terminological systems” and “ques­tion frames.”5 Despite some critiques, the methods of ethno­ science were embraced and used by anthropologists throughout the 1960s and 1970s, and its assumptions framed the work of Sturtevant as a curator of collections at the BAE and after. The BAE was abolished in 1965, and its staff were incorporated into the NMNH’s Department of Anthropology under the Smithsonian Office of Anthropology (SOA). Sturtevant was eventually made curator of the department and became responsible for the collections of North American ethnology. In 1967, he published the Guide to Field Collecting of Ethnographic Specimens, which was revised in 1969. Much like Otis Mason had in his earlier vision, Sturtevant saw the field guide as a way to improve the quality of the specimens collected and the research usefulness of the materials. His guide indirectly pays homage to Mason’s 1875 and Holmes and Mason’s 1902 guides to collecting. The book also provides evidence of how Sturtevant himself sought to classify and arrange collections, despite his claims about classifying knowledge more broadly. It identified what he believed a “good” ethnographic specimen was. The Guide to Field Collecting was geared specifically to museum collectors, non-professional collectors, and trained ethnographers.6 Like Mason, Sturtevant was concerned with the contribution of “data” that would accompany specimens and saw that the lack of catalogue

128

P r a g m at i c C l a s s i f i c at i o n

data (particularly from private collectors) as “the bane of all anthropologists with curatorial or research interests in museum materials.”7 Perhaps the most significant change from earlier eras of collecting was the emphasis on collecting objects that showed “foreign influence”: objects were not dismissed simply because they showed evidence of cross-cultural contact. However, like Mason, he believed that, in order to establish scientifically valuable collections, the most “thorough and exhaustive” collecting was in order.8 Sturtevant provided extensive lists about how to acquire information and documentation regarding terminology, use, and construction. He paid particular attention to developing instructions for the field collector on the labelling of collections, stating that “the best system is to number the items in the order collected,”9 and he provided full and complete descriptions for attaching field labels. The guide even gives detailed packing and shipping instructions. This publication was not for the Smithsonian collections alone: Sturtevant intended it to be distributed and used by many museums and individuals. It seems that his general intention was for people to carry this small guide into the field with them, so as not to miss an important opportunity to collect useful “data” that could transform objects into specimens that could then be offered to a museum for acquisition into a collection. The publication even included extra copies of a “minimum documentation required” pamphlet, presumably so a copy could be extracted from the book and used in context. I do not know the extent to which field guides were used in practice, but Sturtevant’s sits in museum libraries and the personal collections of curators to this day. Sturtevant’s guide presents a hierarchical classification of material culture, which he presents under twenty-one headings, with various subclassifications as well: 1. Manufacture 2. Fire 3. Containers and vessels 4. Transport and associated equipment 5. Killing 6. Gathering for food 7. Animal husbandry 8. Farming 9. Nourishment

The Routine Work of Description After 1950

129

10. Clothing and personal ornament 11. Hygiene, grooming and comfort 12. Medicine 13. Birth, intercourse, death 14. Constructions 15. Games and amusements 16. Music and dance 17. Drama 18. Records, measurements, communication 19. Social and political symbols and devices 20. Religious and ritual objects 21. Trade goods Sturtevant saw the complexities of describing “alternative” or “ethnographic” systems around the world. He advocated that collectors record elaborate inventories and local classifications described in the categories above, in addition to their ethnographic observations, as a way to expend the usefulness of their data set. He remarked that used objects were preferable to new ones, because they could provide “better documentation.”10 De­spite the inability to categorize all material culture in one classification system, Sturtevant felt that his guiding list, much like early circulars produced by the Smithsonian, would be useful in field collecting. He was care­­­­ful to include instructions on documentation and was keenly interested in manu­facturing processes and capturing the entire life cycle of objects as they were created and used. Sturtevant’s guide includes a full section on documentation and the importance of it for maintaining the value of the collection: “It is certainly true that the value of a collection depends almost entirely on the quality of the accompanying data, and that 99 percent of the ethnographic objects now in museums lack essential data.”11 Sturtevant cites another bulletin, “The Object of Research in Museums,” which argues for a “specimen equation” where “Object + Data = Specimen.”12 He explains that data comprise the “notes, measurements, drawings, charts, graphs, photographs and models,”13 which render the object valuable as a specimen. He then supplies a list of what the “documentation required” should be: 1. Field catalog number 2. Ethnic group of users 3. Name of artifact

130

P r a g m at i c C l a s s i f i c at i o n

4. Use 5. History 6. Condition 7. Component materials 8. Typicality 9. Other associated specimens These types of observations are closely related to categories of objects in earlier guides, and many directly relate to the existing categories of collections documentation in the Smithsonian. Sturtevant saw material objects as the documentation of human culture – comparable, he argues, to the “recording, in written form, of data on the nonmaterial aspects of these cultures.”14 Indeed, as he writes in his field guide, artifacts had specific advantages over written works, particularly in discerning behaviour or beliefs, because they were concrete, objective, and “difficult to distort.” For him, they were not subject to personal or ethnocentric bias. Consequently, they were more valuable. As he writes, “Collectors should not forget that many kinds of objects increase in value to their owners with use, because they function better after they are broken-in, or because effective use requires familiarity with the individual peculiarities of a particular sample, or because they acquire sentimental value through long use, or simply through age and associations.”15 Their scientific value, he continues, depends on how the objects were sampled in the field, and the “precision and amount of catalog data” that the collector supplies. His field guide includes a specific section on collecting, in which he plots precisely how to collect these objects and data, or specimens, even without specialized knowledge of the location or community. Like Mason, he sought to fill in the gaps of collecting, and desired the existence of a “union” catalogue of the world whereby all those interested could see, in every museum, what was lacking. He was also aware of possible incommensurability in describing material culture, and he argues that large-scale comparative classifications of objects should not be taken too seriously because of this.16 He suggested that, when collecting, it was more appropriate to rely on local classification, and notes that this typology “would often disagree with the initial perception of the outside collector.”17 He specifically encouraged the collection of Native names relating to objects. Interested in how these objects connected or related to other hierarchies, he gives the example of a screwdriver being considered a member of the category “tools” by “most Americans,” versus a

The Routine Work of Description After 1950

131

“can-opener,” which is considered to be part of a “kitchenware” subcategory. Because of his interest in how people categorized cross-culturally, he urged collectors to note these kinds of relationships as “folk classifications,” which could be compared and studied. These classifications, which were really ad hoc naming standards, were put into practice by staff: they were adopted as naming structures and applied to the entire collection. Sturtevant was also interested in creating standard names for communi­ ties and tribal affiliations, something anthropology had been concerned with for some time. He worked to classify cultural heritage groups by compiling the writings of anthropologists and ethnologists in a series called the Hand­ book of North American Indians (HNAI). The Smithsonian had conceived of the series as a survey of all North American Native communities, customs, and languages, with each section written by a different anthropologist who had worked in the specific area. The volumes are arranged by geographical location – for example, Volume 13 is focused solely on the Plains. The HNAI also includes three descriptive volumes detailing the histories of Indigen­ ous peoples broadly, and the challenges of Indian-white relations.18 The Department of Anthropology at the NMNH used the series to standardize its use of cultural terms, or ethnonyms. When the Department of Anthro­ pology attempted in 1996 to standardize all North American tribal names, the data standards were based primarily on the Handbook of North Amer­ ican Indians series and were sent to Sturtevant and other curators for approval.19 Even in the current National Parks Services (NPS) Museum Hand­­book, the classifications of ethnonyms are derived directly from the HNAI series as well as from the Human Relation Area Files (HRAF).20 Today, the list of ethnonyms developed by the NPS still constitutes the authority file for the subject in the Department of Anthropology.21 These dedicated ethnonym lists, based primarily on the documentation of geographical lo­cations in the HNAI series, prescribe how tribal designation was assigned in collections documentation. In addition to his interest in the standardization of ethnonyms, Sturte­ vant attempted to revise the department’s cataloguing procedures, especially regarding ethnonyms, and to implement his theoretical understandings about material culture within the museum. He was instrumental in creating a list of “World Culture Areas” that was used at the Oklahoma Pilot Inventory con­ference in 1965.22 The primary purpose of this list was to guide the indexing of ethnographic collections around the world. In addition, he suggested a “coding system” for museums to facilitate classification: “This

132

P r a g m at i c C l a s s i f i c at i o n

list derives from the categories in a world-wide general anthropological bibliography of technology and art I have been working on for about ten years ... I am relatively certain and confident that categories like thesauri are an adequate summary of the usual descriptive typology found in anthropological writings – and hence, I presume, also in museum catalogs.”23 In an internal note to the department, he then gave detailed instructions about how an index of material types and descriptions should be used: Instructions: go through this list in this order, entering the object under the first appropriate heading. Where the function or use of the object is uncertain or unknown, skip to the section (17–28) classified by materials and techniques (choosing the dominant or most important one represented in the object). If the function is clear but not here listed, enter the object as 17 other function. If the materials or techniques are mixed, unknown, or not here listed, enter as 30 other or miscellaneous materials or techniques. After coding, if possible add brief (maximum: five words) written description giving more specific details on function (especially) form, or materials.24

In this note, Sturtevant also itemizes the kinds of object types and functions that should be listed, which can be read as precursors to those listed in his Guide to Field Collecting published in 1969. Likely, his curatorial interest in languages and classifications extended into the practice of cataloguing, especially in an early era of digitization as the department was working out how to deal with the new affordances provided by computerized information systems. It is difficult to discern if his “coding system” was ever fully implemented or entered into any kind of proto-database or computerized inventory (like SELGEM), but his desires for this practice is evident in much of his scholarly work.

Changing Work Roles in the 1960s and 1970s Thinking historically about how the NMNH database was actually constructed also highlights the importance of understanding the shift in work roles and in the practice or routine of cataloguing. Since roughly the 1930s, anthropology had become an increasingly institutionalized discipline, where efforts to use material objects as evidence were being replaced by partici­ pant observation, researcher reflexivity, and eventually representation from source communities. In the Department of Anthropology, the care of the

The Routine Work of Description After 1950

133

collection shifted from a traditional reliance on select curators toward a system that included collections managers, data managers, and researchers. Throughout the first half of the twentieth century, the institution’s work roles and routines had stabilized. Discussion of these practices occupied little space in the annual reports. However, during the mid-1960s, an important shift happened. New information technologies were more commonly available, and the Smithsonian was beginning to recognize their potential. The institution saw computerized cataloguing systems as a new tool to increase access to collections and make the information about specimens easier to find. The work of accessioning and cataloguing became more routinized and formalized, and, importantly, it depended on the new information systems themselves. Further, new accessioning guidelines were informed by the mistakes on the catalogue cards from past procedures. A document entitled “Recommendations for Catalog Descriptions” describes in detail how museum staff should record the “complex” and “simple” object descriptions: 1. The description of a complex object should consist of short factual statements, as many as necessary, rather than a single long, involved statement. Each statement should refer to a particular part or feature; and each statement should name the part or feature at or near its beginning. 2. Intuitive interpretations of purpose should be avoided; also conjectures about missing parts. 3. Include, however, statements of purpose, provenience etc. Based on data provided by the collector, published reports, or data provided by knowledgeable consultants. The authority should be cited – either in the individual descriptions if for a few items, or on the history card if for many. 4. Names used at the top of the description should be purely descriptive, and should not include modifiers referring to details of use, age, condition, etc.25

Under this system, the process of accessioning objects created an official written record of the collections, “for legal, reference, and historical purposes.”26 The information recorded during the accessioning procedure was of utmost importance. Unlike in earlier eras of museum cataloguing, where the focus was on the scientific merit of object information, the work of accessioning and cataloguing in the mid-1960s and onward focused on the collections of information for pragmatic reasons. The kinds of associated

134

P r a g m at i c C l a s s i f i c at i o n

information that were required in the documentation of ethnographic materials focused mainly on donorship and acquisition conditions, not necessarily on robust ethnographic information. In the 1970s, when staff accessioned an object, the forms required six fields: Donor of record, Dates, Original acquisition, Provenience data, Identification, and Reference. A memo on cataloguing the ethnological collections noted that the names of the Indigenous creators of objects were almost never available: It would be of value ... if we included information on the name of the Indian creator or manufacturer – where known. However, perhaps we have so few examples where this information is recorded it would hardly pay to include a line for this name in our outline. We do have some, especially works of art, where the Indian artist’s name is known. Perhaps it should appear in the remarks section in every case where the creator has been identified.27

This passage highlights the problem of the lack of information in past records at the same time as it addresses the issue of slotting information into existing data structures. Despite the departmental efforts to improve accessions data, the existing records provided the most confusion and dissatisfaction. Some of the information in the card catalogues had been entered decades earlier, and it was often inconsistent and inaccurate. Despite Sturtevant’s desire and detailed instructions and notes, the curators in particular were frustrated with the state of the descriptions provided on the catalogue cards, both historical and more recent. As Gordon Gibson, a curator in ethnology, complained: I continue to be dissatisfied with the very unsystematic, inconsistent and cum­bersome manner in which ethnological materials are being described for the catalog. Functions, techniques, basic materials, shapes, vernacular names – all appear, singly or in combination, as “name of object.” The descriptions follow no established rules with respect to the order in which features of objects are described, the amount of detail included, or the style of writing. Descriptions are often written in long, grammatically inverted, strings of phrases in which it is impossible to be sure which of several earlier listed features is modified by a characteristic mentioned later. I often must spend several hours reviewing and rewriting the descriptions provided for a collection of newly accessioned African materials, and feel that this is a serious misuse of my time.28

The Routine Work of Description After 1950

135

Saul Riesenberg, also a curator at the institution, suggested that the manual A Pilot Study for Inventorying Ethnological Collections be used as the standard scheme for object descriptions.29 This manual, published in 1967, encapsulates the results of the Oklahoma Pilot Study conducted because of in­ creasing interest, expressed at a number of conferences, in creating a “union catalog” of all of the ethnographic collections in North America (primarily in the United States). Riesenberg specifically argues that the fields established for the Department of Anthropology at the NMNH should reflect the schemes for describing objects as noted in the Pilot Study.30 These were as follows: 1. Object 2. Material 3. Modification technique 4. Function 5. Culture 6. Outline of world culture code 7. Earliest date 8. Catalog number 9. Accession date 10. Specific locale 11. Collector 12. Donor 13. Inventory date 14. Remarks.31 Riesenberg wanted to reduce the “prose” of catalogue records almost completely. As a curator, he had an explicit interest in the organization of the information in the catalogue records themselves. By 1968, there was an inventory of the department’s entire collection; information about the materials, conservation, and storage was collected by small teams as they worked through the storage rooms to assess the condition of the objects. The more “basic” research of “processing, accessioning, cataloging, sorting and maintaining of records” was to be put in the hands of museum aides, technicians, and specialists, with minimal oversight from the curators.32 Thus, the work of cataloguing the collections moved from the responsibilities of the curators toward specialized (and unspecialized) workers: data managers,

136

P r a g m at i c C l a s s i f i c at i o n

collections managers, registrars, and, occasionally, interns. Curators believed that this “simplification” of the process would (in an ideal world) lead to a “more effective and efficient operation.”33 “Freeing” curators from this work allowed them to make identifications and recommendations for the collections as a whole, but it also sidelined the important labour to positions with less stability and power. Hiring a specialist in data processing was not just a hallmark of increasing bureaucracy, but was also required because of the use of new technologies. Understanding objects and collections information as fundamentally important parts of museum work required entirely new museum positions for the purpose of the care of collections information. Despite their importance, these positions were not often taken by field specialists or anthropologists. An acknowledgement that collections information was necessary coincided with a general belief that the actual work of managing these data was unspecialized and “technical.” The care of specimen data was completed by those who were, in the eyes of museum management, unspecialized workers. For many years, the museum categorized these positions as mundane clerical jobs, which often resulted in low pay and insecure contract work – which are still issues today. When the information on the catalogue cards was being translated into the computer database, the work was sent to a data entry firm, further disassociating the practice of collections recording from the perceived intellectual or “expert” knowledge of curatorship. When the museum adopted its first computer cataloguing program, the Department of Anthropology revised the way it handled information about ethnographic objects. This program, SELGEM, or Self Generating Master, was the precursor to other information-management databases used in the institution, such as Inquire and the current system, EMu, a proprietary software system used to manage collections documentation. The retroconversion of earlier collections documentation required that terms be standardized in order to make the system work the way staff needed. The advent of computerization in museums, particularly at the Smithsonian, made it possible to store and retrieve data in a way fundamentally different from what had been possible with typewritten records. As Ross Parry has commented on this history, museums viewed computerization as a “coming together, a drive to standardize and systematize, to make cataloguing more efficient and ordered.”34 There were many attempts to develop a formalized standard of terminologies and taxonomies across museums, including at

The Routine Work of Description After 1950

137

numerous national and international conferences dedicated to this subject throughout the late 1960s and early 1970s.35 Parry argues that the computerization era in museum history and documentation was a “rationalizing discourse,” “aiming to bring order to the bricolage of earlier twentieth century curatorial practice.”36 It sought to simplify and condense hundreds of years of inconsistencies within one system – a formidable task made possible by the introduction of computing technologies.

Computing Culture A return to material culture collecting in anthropology, a focus on collections care within the entire Smithsonian Institution, increased attention to providing access to museum collections, and the development of government computing technologies all led to the creation of an in-house software program to keep track of and document the objects in the NMNH. Although curatorial influence on the catalogue still existed, software development was an institution-wide initiative at the Smithsonian, involving individuals from multiple collections and museums. A scientific staff committee was established to understand the potentialities of data processing in the Smithsonian. This committee worked first with information from the natural historical collections to determine if it was possible to standardize the methods of inputting information. The goal was to develop the first kind of data processing unit as a catalogue for the entire collection, one that used punched paper tape.37 The development of these “automatic data processing programs” required scholarly and community support to establish best practices and guidelines for inputting information and then for searching that museum information usefully (see Figure 15). To coordinate a group of permanent staff members to address issues relating to computerization, the NMNH established the Automatic Data Processing (ADP) unit in 1970 (Figure 16). The ADP helped develop the computer system SELGEM as an in-house, Smithsonian punch-card system. SELGEM held the information from the card catalogues and other related data from the entire institution (see Figure 17). As the user manual explained, “it could handle any type of information about anything.”38 Using SELGEM, the programmer crafted queries onto punch worksheets or punched paper, input these into the system, and then retrieved the information via pages and pages of lists of collections that appropriately fit the specifications in the query. Similar to other early computing systems,

138

P r a g m at i c C l a s s i f i c at i o n

Figure 15  The computer room in the Automatic Data Processing Center, 1978, with the Honeywell mainframe computer model 66/05 and two computer operators, John Spinner and Denise Munns | Smithsonian Institution Archives, Image SIA2010-0870

Figure 16  Tom (Mac) McIntyre catalogues squirrel specimens on a Friden Flexowriter, 1971 | Smithsonian Institution Archives, Image 71-510

The Routine Work of Description After 1950

139

Figure 17  Fred Collier and Jann Thompson looking at the SELGEM printout produced on the Honeywell mainframe computer, 1975 | Smithsonian Institution Archives, Image SIA2013-03885

SELGEM consisted of twenty-five basic programs, which included inputting and editing programs, as well as query, retrieval, and indexing programs.39 More than a collections-documentation program, it was used across the museum and within different departments because of its ability to hold many different kinds of information, from specimen slides to political slogans, mottos, and even banners. Due, in part, to the move to computing across many federal and private industries, programs were developed to house many different kinds of information for different purposes. SELGEM was the first system of its kind and was envisioned, as Ross Parry notes, as a “general purpose program developed for information management.”40 Several professional networks and research groups had developed in the late 1960s to deal with the complex issues that were arising from computerization. The work at the Smithsonian focused on the creation of terminology standards and new information practices and work roles that would eventually be required of many museums worldwide. Standard­ ization and computerization were topics raised at international conferences, and special interest museum groups were established to deal with the increasingly complex issues. Notably, the Museum Computer Network was established in Washington, DC, in 1967, and the Information Retrieval Group

140

P r a g m at i c C l a s s i f i c at i o n

of Museums Association was established in the United Kingdom at the same time. The goals of these professional groups were lofty: to create compatible information structures that could accommodate and exchange the range of data required of cultural heritage institutions. Many of these endeavours focused on data standardization and classification, and the development of terminologies that would work as descriptors for museum collections of all types.41 The advent of computerization in museums (as Parry rightly argues) brought an increased awareness of the organization of information and its importance in facilitating research and providing access to the collections.42 A first step in the data standardization of the computer age was to inventory the collections, in an attempt, once and for all, to establish and revisit the out-of-date and, in many cases, grossly inaccurate card catalogue. The Smith­ sonian saw computerization as a legitimate and pragmatic way to finally solve this issue. In 1977, Vincent Wilcox was appointed as the collections manager for the Processing Laboratory. His duties were to plan new programs and procedures for improved collections management and to coordinate programs between the Processing Lab and Conservation Lab.43 He was responsible, alongside the ADP unit, for coordinating the computerization program in the Department of Anthropology. With Wilcox supervising the data coordinator and the typist, an “Ethnology File” was created from the catalogue cards and typed into the system. During this time of significant change, two issues came to the fore. The first was the standardization of terms in SELGEM, specifically, “object names” and “material types.” With­ out a list of terms on which cataloguers could rely, the department knew computerization would simply recreate the problems of earlier eras, which it was now seeking to resolve. Second, the increased need for a full inventory of the collections became apparent, given that computer systems allowed the institution to gain greater access to its collections information. The inventory issue was particularly pressing because, as a result of longstanding space and structural challenges, the NMNH was due to move its collections from the original building on the National Mall to a new facility in Suitland, Mary­land, which was planned to open in 1982. In anticipation of this, a large-scale inventory was conceived to collect better data about the existing collections and to monitor their condition and space requirements for the move. Both of these projects – the standardization of terms and the inventory – fundamentally restructured the organization of collections information.

The Routine Work of Description After 1950

141

Finding the “Unlocatable” After the catalogue cards had been input into SELGEM, an inventory was coordinated with the ADP unit to allow the museum to check strategically to see exactly what was in the collection, and to begin the process of managing and correcting errors in the catalogue. Staff compiled a number of practical instructions for completing the inventory, and the museum hired a group of people to deal with the creation of the inventory data.44 The main issue was that the museum needed to compile a list of objects “missing” from the collections. Staff completed inventories on specific groups of objects, particularly those that were to be lent to other institutions.45 It was not uncommon that objects were simply “unlocatable” in the data structure or the building. Staff and contract workers created lists of such objects along with the catalogue numbers to which they corresponded. This produced a data file – a file of unfindable objects – that was separate from the catalogue records in SELGEM. After the inventory was completed, the two files – the inventory lists and the collections data – were merged, making it possible to see what was missing and what was lacking correct information, or what was lacking any information at all. The objectives of the inventory-control program and the collectionsmanagement needs of the department were to (1) retrieve data on the collections; (2) plan for collections storage; (3) reorganize the collection; (4) conserve collections; and (5) conduct research.46 As the plan for the collections inventory notes, “The computer at this stage of our development is useful as an indexing tool. It is impractical to try to enter whole information as is done on the catalog cards. The cards are there and available for examination, as are the specimens. The computer does not replace them, but merely provides a quick index to them.”47 Yet, as in earlier collecting eras, the collection of museum information, in addition to ethnographic information, increased the value of the objects as specimens. Wilcox argued strongly to have curators and other staff make good use of the computer, as it would make visible the true scope and nature of the entire collection, ultimately tying the value of the collection to the value of its indexed data: Presently we have no efficient means to define the nature and scope of the collections. The information is there in the catalog, but without the quick multilinear indexing capabilities of the computer, it is difficult to plan systematic collection use according to current approaches in anthropological

142

P r a g m at i c C l a s s i f i c at i o n

research. Once data as to culture, provenience, collector, material, and object name are indexed in the computer, the true scope of the collections can be realized and systematic programs for collection research and use may be planned utilizing the best resources available. Such research provides the basis for future improvement of the database, to be eventually added to the computer within a number of specifically defined research files. The more the collections and the computer system are used, the more effective and valuable they become.48

The inventory of the collections that took place before the move to Suitland was a search-and-find operation to gather information about the collection before it was moved. By 1979, the entire museum had more than 1.5 million specimen records which had been entered into the computer and eight million more needed to be entered.49 The inventory required a physical check of the entire collection, a time-consuming project. The ADP data needed to be compared against the “old record” (the physical card catalogue) to ensure that specimens were all in their correct places. Records were then typed on electronically sensitive paper with an IBM typewriter using an OCR input typing ball.50 They were then entered into an Entrex system using the SELGEM program.51 The data entered into the computer came from two independent sources, the catalogue data and the “specimens themselves.” The data from the card catalogue were entered directly from the cards, and much of the information was copied by non-specialists, who were instructed to simply copy the cards. The other data came from the inventory, and was inscribed on code sheets. This process, described below, involved more collections management work. The catalogue data were indexed according to the following twelve cat­ egories: Catalog number, Accession number, Collector/Donor, Accession date, Acquisition (gift, etc.), Number of specimens, Culture, Country, State, County/Region, Site provenience and number, and Object name. The speci­mens themselves were slotted into eight fields: Catalog number, Object name, Material, Conservation needs, Storage type, Storage index, Status/ Location, and Number of specimens. This museological data served a pragmatic purpose. The information taken from the inventory of the specimens is exactly the kind of data that is useful within a pragmatic knowledge practice at the institution. These are not ethnographic data concerning the nature of the objects and their associated cultural practices and peoples; instead, they relate to their care and collection as specimens or owned

The Routine Work of Description After 1950

143

property within a particular system. Subsequently, additional information was added to the files: the object’s culture or geographic code, its conservation treatment; the negative number of the photo of the object, and the object’s suitability for loan.52 Museum management thought that the accumulation of these types of information would allow for a kind of status report on the objects, clarifying the issues and registration problems that needed to be remedied because of incorrect numbers, miscatalogued objects, and missing specimens. These data were compiled by teams of at least two people who moved through the storage units with a code sheet. The teams quickly jotted down information about each object using eight categories on a machine-readable sheet (Figure 18). This method was preferred to having someone typing up the results, which was regarded as prone to introducing error.53 Each “inventory team” consisted, ideally, of a “caller,” an “entry,” and a “scout”; together, they would make sure that all specimens were entered correctly on the inventory sheet. The department clearly thought the work was repetitive and tedious, and so it requested that students be hired who were looking to gain experience and who had some motivation for doing the job.54 The goal was to complete nearly 750,000 records. Figure 18 shows the Anthropology Inventory Worksheet, which classifies material types into eight major dichotomous types: non-metal, metal, cellulose, protein, resin/wax/rubber, inorganic/organic, synthetic, and elaborate. One could, for example, choose something that was “resin” and “organic” by circling both of the terms on the sheet. There was a single code for each subcategory of material, including combination codes (visible in the third bottom row of the Materials and Conservation section of the worksheet). The completed worksheet functioned much like a Scantron sheet: workers would pencil in the conditions they saw for each specimen and pass these through an optical character-recognition scanner in the computer. Part of the reason why the inventory was undertaken was to prove to Congress that the collections needed better care – which required more funding. As one staff member recounted: “The Smithsonian was arguing, and was ultimately successful, that these collections need better care, they need better facilities, and we’re assembling the documentation telling you of what we need. So it was justification for a certain amount of storage.”55 The museum numbered and correctly identified the entire collection in order to support its claim that the objects needed to be properly cared for and housed. Further, the design of the inventory code sheets was driven by conservation needs in the

144

P r a g m at i c C l a s s i f i c at i o n

Figure 18  Inventory Worksheet, “Prime Record Inventory” | Registrar’s Files, Department of Anthropology Inventory Worksheet, Smithsonian Institution, Museum Support Center

department: as Figure 18 shows, the department’s inventory sheet was dominated by the materials category. As a staff member remembers: The materials, which are quite odd (if you think about it ... from an intellectual, anthropological perspective), if you turn your mind to think about it from a conservation perspective, it was “Oh, well we need to know how many metals, how much material is metal, because it needs different environmental conditions than things that are textiles.” What was in the back of [the conservator’s] mind, of course, was trying to justify our need for a metals conservator, and a textiles conservator. As it played out, we got one environment, period. Everything in the same environment, and we got no conservators to add to our staff. So what was left from that whole endeavor was data on materials that are endlessly confusing to people. Things are made of “cellulose” or “protein fibres” ... It makes sense for the purposes for which it was collected, but it became data that didn’t end up justifying. But we’re left with it. It’s some information; it’s better than nothing.56

The Routine Work of Description After 1950

145

Directions to those doing the coding explicitly noted that the inventory should not record any data considered to be “anthropological”: “The purpose of the inventory is collections management. Qualitative judgments regarding the anthropology and history of the specimens will be the object of other programs, requiring skilled and highly trained personnel in direct collaboration with the curator.”57 Despite this narrow focus, there were some problems with this method. The code sheets implied that any materials circled were “equi-present” in the object. Candace Greene, the collections manager, describes this issue in detail: You’ve got a textile, a cotton, whatever, with a metal button on it. And they’re both noted, equally. So, is it a metal thing with a little string tied onto it, or is it a textile with a little metal thing tied onto it? If you think of it as describing the object, it’s very bad. If you think of it as someone asks a data question that you can’t figure out how to get to through other fields, sometimes they’re like, what do we have? ... I was interested in studying Plains Indian cradles that were a particular kind that didn’t have a wooden frame. And of course, they’re all classified as cradles, so how do you get there? Well, we could do cradles and then we could winnow out anything that has wood.58

In contrast to turn-of-the-century approaches to collections management, the collection of conservation data eclipsed that of the ethnographic “data.” Material combinations – for example, “cellulose fiber and textile” – are still the only material information present in the catalogue today. For example, a wooden rattle was described as a combination of different material types: “Wood/bamboo/cane/reed; dressed skin; resin; shell/pearl/eggshell.” Other objects that included parts such as human hair, were described as being made of “wood/bamboo/cane/reed; protein fibers.” Are these considered sufficient descriptions of the materials in objects today? Likely not. Yet, the inventory code sheets provide evidence of how material forms impose a kind of standardization. Staff used the same worksheets and material types for every object, resulting in a “flattening” of the multitude of combinations that are possible. During the inventory process, it became apparent that an index would be needed for each different category of information. Indices were eventually designed so that it became possible to catalogue the collection by culture, provenance, material, storage type, and object name – fields similar to those designed a century earlier. The computerization of these data made

146

P r a g m at i c C l a s s i f i c at i o n

it possible to search the collection by any of these terms: the storage arrangement was no longer the only index to the objects.59 Multiple indexes were meant to ensure that storage could be better adapted to specimen care, without sacrificing the collection or data control. When the two sources of data – the catalogue cards and inventory index – were later reconciled, existing problems were highlighted, and some of these could not be resolved. This process also allowed researchers for the first time to see where the problems lay, and match the objects with their information.

Pragmatic Classification As it entered the computer era, the NMNH’s Department of Anthropology relied on an organization and classification scheme that it regarded as pragmatic. The department cited cost – including the economic feasibility of entering information – as a reason to avoid storing large amounts of data. For the staff at the time, the inclusion of large amounts of information in the computerized catalogue system was not viable; staff were aware that too much information would lead to “excessive record sizes,” which would be inefficient.60 Owing to technical difficulties and other constraints, including the restrictions placed on large amounts of data, the department needed to determine what constituted the indispensable minimum of information that the catalogues had to include. This decision was complicated because different users – for example, collections management and curatorial research staff – had different needs with respect to the early computer tools: different research questions and institutional uses required different kinds of in­ formation about objects to be recorded in the catalogue. As a result of the limited computerization of data, substantive information about objects existed across multiple sources: the computerized records, the earlier accession files, the physical ledger books, and the catalogue cards. The historical documentation contained information, although the accuracy of this earlier information was always in question. As staff computerized the records, some basic information units – for example, the object type, name, and donor – were included from the earlier card catalogues. However, these were often input by interns or non-specialist workers using their discretion, and mistakes or guesses made in centuries past were directly copied into the new system. Computerization increased the need to begin, at the very least, to truly standardize terms in order to be able to sort things in a search-andfind system.

The Routine Work of Description After 1950

147

The NMNH faced a “universal museum cataloging problem” – the challenge of deciding the information needs of the institution, in addition to the technical limitations imposed by data values and expensive servers.61 This challenge was reminiscent of Otis Mason’s first concerns with the early ledger books – that they were replete with unstandardized terms, making objects nearly impossible to find. As part of the computerization process, staff developed guidelines in the mid-1970s for the data categories in ethnology. These data categories was organized primarily by “major world areas of curatorial interest,” particularly the North American and African collections. As part of this process, staff developed a separate “collector/donor” file for the ethnological collections. Following much debate about how the ethnology collection should be organized for the purposes of the computer catalogue, it was eventually organized by culture region and tribal designation, and eventually by continent and country. At that point, objects were assigned codes in the system solely based on the affiliations noted in the records, not on the objects themselves. The classification of object types into specific categories had always been an issue. William Sturtevant was opposed to any kind of simplified classification, particularly when it came to specific tribes or groups.62 Yet, from a record-keeping and database-management perspective, classification and standardization were essential for retrieval. And so “controlled vocabularies” were used to limit the amount of strict hierarchical classifications but still maintain internal consistency. The department created several such vocabularies in an attempt to standardize the variable naming strategies at the institution, a distinctly pragmatic approach to the management of cultural objects. In a good description of their function in institutions, librarian Ann Doyle notes that controlled vocabularies perform three functions to disambiguate meaning: the control of terms to prevent multiple terms for the same concept (synonyms); the control of terms to prevent multiple meanings for the same word (homonyms); and the mapping of variants in order to show relationships. The mappings show structured semantic relationships, such as hierarchical, equivalency, and associative relationships, and perform a navigation function in showing users relational knowledge.63

The controlled vocabularies – index terms, geographical names, culture names, and people names – become information once linked in the catalogue

148

P r a g m at i c C l a s s i f i c at i o n

record. The Smithsonian’s catalogue records similarly perform relational knowledge: they were inscribed versions of the hierarchical and irrefutable connections between donor, land, and property. The index terms were the controlled vocabulary or nomenclature of the anthropology department. These terms were imagined as a way to enable the sorting of information about the catalogue in the computerized files: “Index terms provide the initial point of entry into the ethnology collation of the Department of Anthropology in the National Museum of Natural History. They are not intended as a hierarchical taxonomy or typology of artifacts, but are rather a guide to research in the collection.”64 Staff were aware that a true standardization of terms would not, and should not, be possible. Even one hundred years earlier, Otis Mason, who was dedicated to the classificatory concepts and organizational concerns of the institution, noted that no single nomenclature could ever suffice in the description of objects: “There is a danger of overloading the subject with too many difficult names lending rather to confusion than to perspicuity.”65 Where Mason was concerned with the kinds of information and the physical and theoretical classification of the objects in the collection, the era of early computerization in the department made other kinds of comparison necessary – those required by data standardization – so that objects could be intelligible to researchers and staff. Computerization made it possible to sort extremely large amounts of information, which made the ethnographic collection more accessible than ever before. Attempts to create standards of object description resulted in the creation of subject-specific thesauri and dictionaries, and these were translated into the data structure of the early computing systems, and eventually the modern database. The creation of the fields, and the standard ways to fill them with information, became the central focus of the collections management division of the anthropology department for some time. A certain curatorial disdain for the computerized system of records perhaps motivated the introduction of standardization as well. As Gordon Gibson, a curator in ethnology at the time, noted: “I presume it is important before starting to computerize the ethnological collections that we agree upon standard terminologies to be employed in cataloguing. We have long needed to standardize, and the greatest benefit of computerization may be that it now will force us to do so.”66 It was conceived that a “ready-made list” be adopted to document cultural affiliation. Gibson proposed that the department use the HRAF codes in their digitized files as supplements

The Routine Work of Description After 1950

149

to other culture-area locative data, opining that there was “no better standard system.”67 Gibson noted that the HRAF’s hierarchical order from culture area to culture to subculture was useful, but proposed that it should be “subject to modification.” However, to include a hierarchical order, a “master file” of culture names and synonyms would need to be created and maintained – a massive undertaking. With the advent of later, more robust computerized systems, the most significant issues facing the digitization of information about objects and records in the department were standardizing index terms; determining locality and establishing tribal nomenclatures; and finding a way to correct accumulated mistakes resulting from extensive backlogs in object documentation. Without index lists of preferred terms, it was impossible to know if a query submitted to the database would return all of the results, making it impossible to tell what exactly was in the collection.68 Standardizing names of index terms, as well as ethnonyms and geographical areas, would have a significant impact on the department’s ability to respond to requests from researchers, plan exhibits, organize storage, and plan collections development. Eventually, too, such standardization would affect the museum’s ability to handle repatriation requests: without access to accurate lists of collections, communities could never be sure that they would access objects needed, or that all of the objects requested would be found.

A History of Naming How does one classify and document culture – something that is shared by individuals, through language, history, and protocol, but that changes through time? The preoccupation with such classification was one of De­ partment of Anthropology’s earliest concerns, yet classifying communities and heritages has often been harmful.69 In one respect, the motivation to standardize the ethnographic collections at the Smithsonian came out of an era that was dedicated to making computer systems work. In another, it grew out of an increasing and fundamental frustration with the system of collections documentation that some members of the department felt to be in complete crisis.70 The catalogue data were reportedly difficult to input into a computerized system for several reasons. First, many terms were used to describe the objects, and there was no way that a computer program could automatically detect which terms were synonyms for each other and which were distinct. Object names were ascribed by whoever had catalogued

150

P r a g m at i c C l a s s i f i c at i o n

historically, and the level and detail of description for these names often depended on the role of the individual who had catalogued them and the availability and completeness of documentation associated with the accession or object. Despite earlier attempts to create functional nomenclatures for the collection, the terms that were originally inscribed onto the catalogue cards as part of the system of knowledge were, for the most part, not mutually consistent, which meant that various terms had been used to describe the “same” kinds of objects based on the observations of collectors, curators, or museum assistants; moreover, these terms were often incorrect or out of date. Issues arose when terms that had different meanings functioned as synonyms and had been employed haphazardly. If a researcher searched the computerized catalogue using the term “lance,” all of the harpoons and spears could be excluded from the results, depending on how these objects had been entered into the system.71 These computerized sorts worked only as well as the physical system of inventory, and, even though one could search across data fields, the fields first had to be defined in a meaningful way. The use of a thesaurus and the development of index terms solved this problem. They allowed the museum staff to develop a set of terms that were broader, and therefore fewer in number, with subheadings for various classifications. These broader, more standard terms would be inexorably linked to the object record, but still allow the object name to be recorded in a separate, primary field. For example, these defined and standardized index terms allowed researchers and workers to search for “carving” and retrieve all objects that were previously defined as “wood carving” or “carving, wood.” It is crucial to remember that the source of all of the information relating to the provenance of the objects came from the catalogue cards, the ledger books, and, in some cases, the accessions files. This information was the basis of the catalogue record and that was entered into the computer systems. By 1987, the “primary goal” of the department was to standardize the terminologies used in the system, yet several vocabulary-standardization attempts presented recurring problems. In 1988, a sample set of 12 percent of the collection showed that there were 1,436 separate terms used to describe the objects.72 Internal memos and documents highlighted the need to correct a variety of errors that resulted from the inconsistent conventions of object-naming terms. For example, the same types of objects sometimes received radically different index names; different spellings of the same term also created challenges. A 1987 memo recommends the use of singular nouns for objects (e.g., “moccasin” instead of “moccasins”), and that multiple

The Routine Work of Description After 1950

151

variant names be subsumed under preferred common names (e.g., “bag” instead of “pouch” and “sack”). The memo recommends the use of individual rather than collective names (e.g., “shirt” and “leggings” rather than “clothing,” although “clothing” would be used as a “functional heading” under which related objects would be gathered). It also notes that descriptive adjectives or construction material were to be ignored. Further, functional criteria were to be “emphasized wherever possible”: the memo gives the example of using the term “pipe bag” for an object earlier catalogued as “bag, pipe”).73 Index terms were added as a workaround to the problem of inconsistent object names that would not function well for the computer sorts that were needed. Alternative names were added in a separate field, which explains how updated naming practices and historical ones existed together: for instance, a “rattle” was also termed a “carved wooden rattle” in a separate field. A list established in 1987, marked “Preferred Index Specimen Names,”74 established a set of rules for the creation of the thesaurus. Descriptive adjectives and construction material were ignored in specimen names, although they could appear in subcategories of objects in other catalogue fields. This approach is visible in a draft of the structure of the preferred names for index terms (Figure 19). The goal of standardization was to reduce the number of synonymous terms and the redundancy in names for objects by further reducing the number of preferred terms for index names. For example, by 1988, there were more than 62,000 objects in the North American Ethnology collection, which were catalogued using 9,918 terms; by comparison, institutions like Pitt Rivers used fewer than 1,200 terms.75 Although collections staff were directed to keep terms “simple and common sense,” and to give preference to using single words and to avoid multiple terms,76 examples of increasingly confusing data were common. One document contains a list of objects that had “strange” data, and these were entered into a “Whatsit” file. The “Whatsit” file contained records with index data, and staff were required to cross-reference this file with the object in storage to determine the correct information.77 This list contained records that had a vague index name, and that could not be “reconciled” by going back to the catalogue card and examining the data. Once all of the information had been entered correctly, SELGEM’s functionality allowed for the production of a self-generating list of terms in the collection. Staff then recorded all the object terms in the

152

P r a g m at i c C l a s s i f i c at i o n

Figure 19  Excerpt from preferred index specimen names, memo from Thomas Kavanagh to Candace Greene, 2 February 1987 | Department of Anthropology, Smithsonian Institution, Museum Support Center

system, examined these to select synonyms, and established preferred terms to indicate proper broad and narrow relationships. The index terms were really a kind of inductive thesaurus. The terms were chosen first by those working closest with the collection – the collection management team – and then were passed on for curatorial review. Major headings appear to be modelled on or, at the very least, compared with those at the Pitt Rivers Museum. In at least one case, the department hired a specific individual to do the work of standardization – that is, to review the existing terms and

The Routine Work of Description After 1950

153

determine whether any of them needed changing, and to provide a guide to the index indicating preferred terms and why they were selected.78 Index terms were explicitly designed not to be strict typological classifications, and it was never the intention to make these terms universal or exclusive. The department struggled with how to make the system general and consistent enough but still preserve inconsistent yet historically informative data. Yet, differences were masked and occasionally ignored in order to allow for a sufficiently simple system. The intention was to provide the researcher greater access to the collection. For example, if all scrapers, fleshers, and beamers were subsumed under one heading, “scrapers,” the researcher would need only that term to find anything similar to scrapers, which would perhaps shed new light on the researcher’s interests. Terminologies were influenced by several different sources, but, to this day, there is no formal North American standard for ethnological index terms.79 The department used terms that evolved directly out of the content of the collection, the attitudes and knowledge of the curators, and a desire to maintain the collections and their records as pragmatically as possible. The development and management of the computerized system required much work on behalf of the staff and resulted in an overabundance of terms, with the attendant impracticality. Above all, index terms were intended to be finding aids, a new concept that arose with the advent and greater adoption of computerized sorts and that is addressed by Candace Greene, in a memo to staff in June 1996: The index term has been a useful tool ... However, I consider it a flawed system, and it seems appropriate to comment on its limitations at the same time that I am circulating these guidelines describing the very nebulous rules upon which it operates. First, it was created with almost exclusive reliance upon the object name as entered into the database. These names are often ambiguous or even misleading and the index cannot improve upon that without individual object research. Second, our data structure makes no provision for a hierarchical classification. Therefore, a choice always had to be made between grouping things under larger functional terms or retaining more specific, meaningful terms and increasing the number of entries under which one must search to be certain of catching all possible examples of an object type. The choices that were made are largely arbitrary and are based in part on the number of items within a given category and in part on awareness of how the collection is used.80

154

P r a g m at i c C l a s s i f i c at i o n

This passage highlights the key issues with index term standardization that still exist today: the dependence on the object name as entered from the card catalogue, the lack of a hierarchical classification, and the arbitrary distinctions made between functional object types. The lack of a hierarchical classification system made it hard to determine where objects should fit, and, without such a system, it would be hard to provide consistency in naming through time. Even if this system were inherently incorrect and did not represent any local ways of knowing these objects, it would at least have been consistent enough to provide some kind of regular system of naming. However, the use of index terms as broad headings enabled staff to retain the historical information within the “object name” category – essentially a direct representation of the original name on the object’s catalogue card. This is a squarely pragmatic approach to understanding object relations, which was virtually required when using early computers as sorting tools. It is influenced by the mode of research in the department, which relied heavily on the documentation about the objects and not the objects themselves. The material documentation thus became a site of negotiation. New ways of organizing and standardizing resulted in the application of new terms to objects, and occasionally this was done in sweeping global changes applied to all objects of the same type without reviewing each individual record. The documentation about the objects came to stand in for the material things in the collection.

Shifts in Documentation Practice Much of this book has been dedicated to understanding the boundaries of museum bureaucracy, and how this changes over time. What was an ethnographic specimen in 1965, and why does this matter? It was an object, a knowledge-object, a kind-of-thing to be managed, catalogued, and potentially accessed. It was also an outcome of ethnographic fieldwork, where objects were both carefully defined and haphazardly collected. Effectively, asking what a specimen is begs the question of how a specimen becomes. Specimens are made through their collection, description, and documentation in a system – the catalogue – that precludes particular ways of knowing. As I have shown, these organization devices changed through time, and we can question if Otis Mason’s dreams of a universal ordering device took hold a hundred years later. As William Sturtevant declared in the 1970s, arti­­ facts were concrete objects of culture, difficult to distort. Shifting epistemic

The Routine Work of Description After 1950

155

commitments to understanding ethnographic research and the role of material culture played a major part in how the Department of Anthropology saw its collections and what was needed to preserve them over time. Sturtevant had a great influence on the study of material culture, which in turn influenced (directly and indirectly) the care of the ethnographic collections in the department, and the objectivity of anthropology once again focused on the collection, distribution, and documentation of material things. In the 1970s, computerization required “pragmatic” classifications and focused on how to organize collections in physical space. The development of an in-house computing program, SELGEM, and the move to a new building affected how the computerized collection was organized. At the same time, staff roles changed. The need for data managers, inventory specialists, and collections managers grew, but their roles were often defined as not especially intellectual – unlike the curatorial positions. Yet, they were the ones responsible for organizing the collection and its information on a day-to-day basis and for inputting the records into SELGEM. Like in earlier decades, their work was subsumed into the architecture and data structures of the museum and rarely publicized. Making the system work was the primary goal of the institution, and this is reflected in the way collections staff dealt with the increasing capabilities of computing. Since this time, staff and researchers have continued to encounter the hereditary language of centuries past in the catalogue cards and within computer documentation.

156

P r a g m at i c C l a s s i f i c at i o n

5

Object, Specimen, Data

Computerization and the Legacy of Dirty Data

GENERAL INVENTORY PROCEDURE Please do not try for a speedy rather than accurate inventory. If you have any questions, no matter how trivial, please stop and ask for help from the Processing Lab. If you get tired or confused, stop and rest, when mistakes are made, it may take another hundred years to find and correct them ... The nature of the work is repetitive, and would be tedious for the advanced professional. With tedium comes error, and this is to be avoided. – Robert Elder, Report on the Cheyenne-Arapaho Inventory Project, 1978

In their work on classification, Geoffrey Bowker and Susan Leigh Star argue that the explosion in classificatory principles and thinking in the late nineteenth century acted as both political forces and “organizing rubrics for complex bureaucracies.”1 In the previous chapter, I traced how a pragmatic classification became the new approach to ordering rhetoric of the Department of Anthropology of the National Museum of Natural Hist­ ory (NMNH), in part because of the development of a new computing system: SELGEM. Making visible the way that SELGEM was organized is key to understanding how knowledge was produced within the museum bureaucracy more broadly: within ledgers, card catalogues, or other, newer

computer-based systems. These structures are strongly historical. They are dependent on technological advances, routines, and the development of new standards in making data accessible. As Martha Lampland and Susan Leigh Star have shown, information infrastructures like these need classification schemes and standardized documentation, but they are also practices that are part of everyday work.2 Some of these classifications are formal; some are not. While formal standards are present in the anthropology department in the form of digital standards for information sharing protocols, such as file names, for the most part the department relies on a set of informal standards to conduct cataloguing practice, and has done so throughout its history. These informal standards take shape as continually updated controlled vocabularies, ad hoc naming conventions, lookup lists, and inventory control terms, some of which I reviewed in the last chapter. Naming is a world-building activity, and naming practices are central to how we imagine the iterative, mundane ways in which information becomes knowledge. Over much of the history of museums, primarily white, male curators with similar education backgrounds have made decisions about objects that come from thousands of different communities and geographic locales. The categorizations of material culture are rooted in the values that these curators have brought to their positions. Yet Indigenous peoples have always been active participants in constructing and reconstructing their heritage for settlers and colonial states, and museums have changed substantially in response. There are specific historical moments that we can examine to show how museums – particularly, in this case, the Smithsonian – organized and re-organized their internal logics according to epistemologies or systems of value (whether these were explicit or not). In the United States, the advent of federal repatriation legislation is one of these key moments that brings the history of documentation and the struggles to standardize for new computing environments into a new light. Indigenous communities have been working with collectors and ethnol­ ogists (and as anthropologists themselves) for many years, and in many cases have requested their objects back from museums. Requests began at a federal level in the United States in the late 1980s, when the National Museum of the American Indian Act (NMAIA) and the Native American Graves Pro­ tection and Repatriation Act (NAGPRA) were passed as law, in 1989 and 1990, respectively.3 This legislation required federally funded institutions to create full lists of ethnographic objects in their collections that came from any federally recognized Native American tribes in the United States.

158

O b j e c t, S p e c i m e n , D ata

Through the process of making these lists, staff occasionally encountered information that was inaccurate or misrepresentative, and occasionally did work with tribal leaders to address this. As the political landscape changed, and as Indigenous peoples’ voices for their rights were increasingly recognized, information attributed to objects in earlier epistemologies was, at best, out of date and, at worst, racist and harmful. Such information is a kind of colonial “legacy data” that stick to records despite decades of institutional change. It is a remnant of the colonial past made material in technologies and practices. Working with, viewing, and repeating these offensive terminologies or systems of value in contemporary museum work is a reperformance of the colonial encounter. This observation does not necessarily mean that this information should be effectively erased, and, indeed, this re-performance may be necessary in order to point to and continually question these colonial categories, which otherwise may become completely erased and historically invisible. Yet the influence of the NMAIA and NAGPRA on the attribution and standardization of tribal affiliation in collections documentation cannot be understated. To understand why and how legacy data have become important for federal repatriation requests, a brief review of the contemporary digital catalogue, its guiding standards, and its online presence is necessary.

Filling in Boxes Databases, which support most museum cataloguing systems, use international standards as well as information standards and what Lampland and Star call “rules of thumb,”4 some of which I have explored in this book. These standards are often taken for granted and are implicit in the system, because working infrastructures are often invisible. Yet, in the Department of Anthropology, the individuals who use the infrastructural system are keenly aware of its limitations and of the theoretical implications of these limitations. Several people I spoke with talked about the ineffectiveness of using “boxes” (categories in the database) to hold complex and robust information about ethnographic collections. However, many of these individuals acknowledged that these boxes were still needed in order for the database to work. The Smithsonian and the NMNH have a variety of practices and methods for recording information that differ across departments and among individual curators. The computer catalogue has had three major versions:

C o m p u t e r i z at i o n a n d t h e L e g a c y o f D i rt y D ata

159

SELGEM, the first digital documentation system in the late 1970s; its successor, INQUIRE; and EMu, the current system. EMu is a content management system (CMS) based on an object-oriented database protocol that can manage a wide variety of data. Although the CMS is tailored for museums, it complies with other standards and reference models as part of the data architecture, which includes the ability to comply with standards for exchanging information, metadata standards, and information retrieval protocols.5 Although it adheres to common standards, EMu accommodates site-specific best practices, and individual museums still provide “userdefined thesauri.” At the NMNH, each department has its own nomenclatures, what might be called “ad hoc” standards.6 Although the Department of Anthropology has no truly formal standards for terminologies or data entry, a complex system has evolved through the years as new procedures have taken hold, and as inventories were completed. As I have tried to lay bare, the ad hoc description of specimens in the department during computerization developed out of a history of recording objects that, at its core, is the embedded accumulation of changing ideologies since the late nineteenth century. Today, EMu includes a separate module for the anthropology “catalogue.” This catalogue is a subset of a much larger database, which includes other general functionalities for recording information that are distinct from the individual object records. For example, the “locations” and “collection events” fields can have multiple values. Policies have been developed to ensure that staff follow guidelines and they are encouraged to enhance catalogue data in their daily work.7 The “Anthropology Data Manual” (2012) is the current data manual in use in the department today, and it maps the EMu database and all of the possible fields used to describe the anthropological collections. EMu is made up of multiple modules, and in each module there are several different fields or tabs that can be populated. These tabs contain all of the catalogue data for the objects, including the images of catalogue cards, original ledger pages, and images of the objects themselves. The database itself is object-centred and involves the catalogue record as the central module, with series of supporting modules built around it.8 Each field is a place for one specific kind of information; the field is then organized into groups, and each field has specific value limits defined by the developers of the system. For example, the field “catalog number” is a numeric single-value field, whereas “collection name” is a text field with

160

O b j e c t, S p e c i m e n , D ata

multiple values. Each record has an additional number assigned to it – called the “internal record number” (IRN) – which acts as an internal unique identifier and does not replace the catalogue number. The system allows for Boolean radio buttons defined as “yes,” “no,” and “blank.” Each record also has a multitude of different fields and field groups attached to it. In the example of an object’s cultural affiliation in the manual the field “culture/ ethnicity” has a current cultural affiliation but also secondary historical affiliations.9 This approach denotes that a particular object has had other cultural affiliations that are not static, and the museum recognizes that the assigned affiliation can change depending on the kinds of evidence available through time. Cultural affiliation is perhaps the most important category in the database, particularly for repatriation cases today. Because of the way information was collected in the past, and perhaps because of the paucity of such information in the collection’s records, many Indigenous community members who wish to view object collections begin their searches with the location where the object was found, or what tribe it is associated with. In current collaborative web-based research projects, considering how and with what terms Indigenous people themselves would like to search is an essential part of providing respectful access to collections and other kinds of intellectual property. When collections are researched anew, occasionally there is new information that can be attached or included in records. With the help of community researchers or experts, curators may make recommendations and grant approvals on what changes are needed, and staff members are responsible for entering this information with any relevant notes, names, and dates. Because of the size of the collections, the majority of records do not contain such additions. Attributing names to documentation potentially allows others to trace the intellectual roots of thought processes, and occasionally to trace mistakes. Knowing the name of the person who catalogued the object could help others who use the information to assess its validity. Such an approach is an inversion of the earlier systems that used information taken from the collector uncontested and at face value. Associating named individuals with information in the catalogue is useful, and it opens up museum staff files to scrutiny more readily. Until 2003, there was no place to record the name or date of the person who made the attribution; so that information is often not present in the vast majority of records and is not available publicly.

C o m p u t e r i z at i o n a n d t h e L e g a c y o f D i rt y D ata

161

Several authority files act as standard data values to be entered into cer­ tain fields. In EMu, the authority files are used as controlled vocabularies. Using controlled vocabularies ensures that there are no misspellings and, as the department’s data manual asserts, allows “faster and more accurate data entry” and “more reliable searches.”10 An authority file is supported by what is termed a “lookup list,” a field with a drop-down list of all possible values. Lookup lists are akin to the lists of index terms that were once printed by hand and used by registrars and collections managers (and, to some extent, still are used similarly today). Hierarchies link separate lookup lists from different fields. In the anthropology database, there are hundreds of lookup lists; only a few have truly controlled vocabularies. For Locality fields, the controlled vocabularies exist for Continent and Country. The other fields (State, City, Site name, etc.) all have lookup lists, but they can be edited by users, so they are not truly “controlled” – this lack of strict control allows for greater ease of data entry and helps to reduce spelling variations. With these controlled fields, staff members (with proper permissions) can add new values to these controlled vocabularies as they wish. In this way, the data lookup list is a kind of data entry shortcut. There are other fields as well – Index term, Culture (a set of four fields, from general culture area to specific communities), Material types, and Storage location. The catalogue module in the current EMu system contains twenty-five different tabs and hundreds of different possible fields for information, with around forty of these being the primary or core fields. The data fields that govern the organization of objects – the field Index terms first used in the previous database, INQUIRE – have a controlled vocabulary of 4,462 distinct terms. Some of these have multiple levels of specificity; some have “legacy data” changes that resulted from when individuals changed terms in the database or added pre-existing words to objects. In one example, the “lookup list” for “Rattle” in the ethnology catalogue states that there are 933 “Rattles” and 1 “Rattle/Bag” and 2 “Rattle/Models” in the database. This classification means that more than one object is grouped in the same catalogue record, and so both terms are separated by a slash. The lookup list is another representation of how terms have changed through time. It demonstrates the inherent flexibility of the system, which allows for index terms to be continually added, an effect of infrastructural systems being both durable and flexible. The creation of these controlled vocabularies was the result of efforts in the late 1960s and early 1970s, but the backlog of cataloguing objects stretched years, and, pragmatically speaking, specimens

162

O b j e c t, S p e c i m e n , D ata

needed to be recatalogued in terms of “modern classifications” in order to facilitate scientific study.11 Certain fields are more formally standardized, in the sense that users cannot edit them as they wish: Material type, Culture, Continent, Country, and several internal fields like Storage area code. The field Material type is standardized based on a set of materials that are broad categories that were assigned during the 1970s inventory.12 The Culture field includes the ethnonyms – the standardization of which raised concerns early in this century – which are not editable by users. These fields have been created specifically for the computerized catalogue, with the sole function of facilitating searches. The Index term field is semi-standardized, meaning that staff can enter new terms. As explored in the preceding chapter, it is used for types of objects, and the terms are often the same ones that were used during the creation of INQUIRE and the standardization of search terms in the mid-1990s. Although the department acknowledged that these fields were useful for researchers, it considered them to be simplified and subjective terms based on the “object name” field from the catalogue cards and therefore not accurate identifications.13 Over the years, the department had constructed several “lists” for all three fields, yet there is still no truly standardized list in the database, and these fields were not predefined when information was entered into a catalogue record. This is because the objects in the ethnology collections resist any clean cataloguing naming strategies, in particular because of their multiple meanings in different contexts, their flexibility to change meanings through time, and their different ontological realities in different cultural locales. Indeed, in contemporary research-based collaborative documentation projects that use database structures to organize information, objects are almost always associated with a particular geographical setting and are smaller subsets of collections. Working with communities or tribes in these cases has made us aware of different, culturally specific, modes of organization: objects may be more correctly grouped by their use or their ancestral ownership, for example, rather than by their functional names.14

Data Histories: Transfer and Migration through Time After nearly 160 years of ledgers, card catalogues, and translations between documentation types, much information about the collections has been either lost or unverifiable, or is impossible to reconcile with current political climates and with the source communities that have increasingly been gaining

C o m p u t e r i z at i o n a n d t h e L e g a c y o f D i rt y D ata

163

access to their collections, which had been displaced for centuries. The transfer of terms and information from the paper ledgers and the card catalogue also shows how descriptions become more durable once made material. In the 1960s, the catalogue cards were copied directly into the first data­base, SELGEM. Some viewed this part of computerization as “frustrating” and “time-consuming,” but it was an incredibly important process and seems largely to have gone unrecognized.15 The information on a catalogue card was manually entered into the system as an individual record, but frequently the information from the cards was simply never input into the system be­cause of time pressures. The use of various names (for example, the ethnonyms defined in the Handbook of North American Indians, as dis­cussed in Chapter 4) and alternative names for objects caused increased confusion because, in the search-and-find inventory system, if one did not already know the terms to look for, the information about the object could never be found, a kind of loss or forgetting from the collection directly imported from the cards. When the information from the catalogue cards was entered into the computerized system, staff physically marked on the cards in red what parts and pieces of information they would keep. Card records show that geographic places and object names were often underscored, delineating the information from the older card catalogue that would be entered into the system. Figure 20 shows the traces of this practice as the catalogue cards were physically modified for the newer system. Today, all of the terms are included as alternate terms, so a user can conduct a search with any of these terms and does not need to guess which one is correct. Further, staff prioritized creating a basic version of all the records over creating a complete and vetted sample of each record, with the expectation that serious researchers would be able to get a general idea of what collections existed and were of interest. Robert Elder, the collections manager throughout the 1970s, expressed his frustrations with the underlined cards: “As a general rule the entries faithfully reproduce the items underlined on the cards, but ... I am at a loss to explain why certain items were underlined in the first place.”16 After the cards were underlined, they were sent to a data entry firm. The red marks were taken as authoritative, as the employees at the data entry firm were not subject specialists, but it is unclear who was responsible for deciding on what terms to underline, and therefore input, in the first place.17 If the underlining was incorrect or misrepresented any of the original catalogue card data, it would compromise the ability of curators and collection managers to interact

164

O b j e c t, S p e c i m e n , D ata

Figure 20  Example of catalogue card marked for defining terms for the computerization of the record for a carved rattle | Department of Anthropology, Smithsonian Institution, Museum Support Center, E73790

with the collections. Many exchanges between curators and collections managers focused on frustration with the level of care and documentation that existed in the records that were input from the card catalogue, but with­ out this data the collections would have been impossible to search. The data were not always stable, nor did they always fit within the self-described logical system the museum sought to create. The marks on the catalogue cards are traces of decision-making processes and of the revision of terms that should be put into the system. The invisibility of the infrastructure of the department and the museum is visible in these moments when things fail to work, or when staff members encounter legacy data in the catalogue. It highlights the key point I have explored about the nature of these documentation media – that they are temporal and performative. A memo from William Fitzhugh to John Ewers (both curators at the institution) describes the process of computerizing the catalogue cards and elucidates the differential needs of the archaeology and anthropology collections. Fitzhugh notes how past inconsistencies in record keeping affected the then current records in the context of provenance: A goodly proportion of the specimens in ethnology cannot be clearly documented by name of field collector or date of field collection; while some of

C o m p u t e r i z at i o n a n d t h e L e g a c y o f D i rt y D ata

165

those known to have been collected by a particular individual on a particular reservation may not be creations of the Indians who lived on that reservation, or their ancestors. To be most useful our records should reveal degrees of certainty of the tribal identification of the objects. It would be creating future problems if inexperienced persons were to try to judge of the completeness and accuracy of our catalog record.18

Because of the issues and the cost associated with repairing the computerized catalogue records, the department proceeded more slowly when recording information. The hope was that, instead of simply recording the text on the catalogue cards, staff could return to the collections themselves and mark each catalogue card as each individual specimen was considered. Yet this ideal was unrealized. Fitzhugh’s recommendations demonstrate an assumed trust that the object contains information – that at least part of the value of the object is inherent in its form. The department established general principles for the rules of data entry when the data from the card catalogues were entered into the new system. No changes were to be made on the catalogue cards, and only the information that was on the cards was to be entered into the computer, without any revision. George Gibson suggested that “spellings of names of objects, cultures, and places will be exactly as shown on the catalog cards.”19 If two or more object names appeared on the card, the names were all entered into the computer, with the “common English name” as the principal name. The information about the locality (place made, place collected) was to be entered with “up to three levels of specificity,” with a North American focus (country, state, town). This resulted in inevitable spelling variations and mistakes, which caused innumerable issues when working with SELGEM. Computerization made it possible to search large amounts of data, but for the most part it directly adopted the strategies implemented by the earlier catalogue cards. Uncovering and working to fix the earliest issues relating to cataloguing material culture – for example, the fact that object names could not always actually be recorded exactly as they had been written, and the use of alternative names for objects – was a part of the invisible work of staff. Yet, because of the way that collections information was digitized, staff could not always determine which object names would be ineffective as search tools.20 The “red marks” and the notations that denote the kinds of data to enter were inconsistent at best, and the information culled from the cards in the computer system presented a genuine problem

166

O b j e c t, S p e c i m e n , D ata

for collections managers and researchers who wanted greater access to information about the collections.21 Many people worked to improve the fields, add new ones, and make the data as accessible as possible. For over twenty years, the work of those who deal with the digitized catalogue records today has involved parsing these original catalogue card records into new fields and adding data from outside sources like visiting researchers and community members.

“Bad” and “Dirty” Data Some of the early cataloguers and collections management staff at the museum found the computer system created to deal with the inventory and collections management “of minimal use” and rarely reliable, although this may largely have been a result of lack of interest or ability in learning to use a new system.22 The majority of the work in collections management after the inventory and the move to the Museum Support Center was dedicated to remedying cataloguing errors and evaluating the state and variety of object names that existed in the merged inventory and catalogue data. The need to make terms standard so that they could be matched with one another, and to turn the data on the catalogue cards into machine-readable format, required curators and researchers to evaluate the state of the objects in the collection. Through such a process, they sought to determine where the errors lay, and what exactly they were. One significant and recurring challenge with respect to data entry of the inventory was that the nature of the work was tedious. As Wilcox notes in his report for 1979/80, “numerous typists have revolved through the unit due to the boring and repetitive nature of the work.”23 Despite the staff turnover and a lack of incoming staff, the department purchased new computing terminals (an Entrex 680 and a QUME terminal) in order to complete the computer cataloguing work. By 1980, the ethnology files on Oceania, Asia, Africa, and North America had been “cleaned.” The number of naming categories was expanded as the corrections were made to the files. The inventory had raised issues about data mismatches and the issue of legacy data in the records. To address the potential mismatch of objects, staff checked code sheets against the data entry for typographical errors, which were corrected, if possible, or checked the objects themselves against inventory data. Another problem is that objects in the collection had missing data. If no corresponding catalogue card existed for an object, the staff

C o m p u t e r i z at i o n a n d t h e L e g a c y o f D i rt y D ata

167

turned to the original ledgers to attempt to retrieve a match, yet the ledgers themselves proved incomplete in many cases as well. Finally, once all other sources of errors were checked (the inventory data, code sheet, and ledgers), the object was checked for a corresponding catalogue number. If objects were mismarked or incorrectly numbered, it was the responsibility of the processing lab to remove old numbers and retag them. Missing items in the anthropology collection totalled 3,600 objects, or 10 percent of the entire collection. Perhaps because of these gross inconsistencies, the department drafted the report “Projected Computer Programming Support for Collec­ tions Man­agement,” which outlined the advantages of developing a way to sort the catalogue file in tandem with the inventory file.24 The challenge was to sort the inventory data according to the categories in the catalogue file, such as culture and provenance, which would provide the possibility to sort the catalogue across the inventory file. These “dirty” files were corrected by the end of the fiscal year 1982. Correlating the records was a frustrating and time-consuming task, but through it many objects were “found” within the collections that may have been otherwise lost. This work required meticulous attention to detail, highly developed skills, and a deep knowledge of the collections – knowledge work that has long been invisible within the museum. Despite the challenges, this work made it possible for researchers and other people to ultimately find what was in the collection, itself an important reconciliatory project. This work was still in progress in 1982, and the data that had been entered by the outside firm in the process of computerizing the card catalogue continued to be “proofed and standardized” and compared against the results from the inventory.25 At this time, the computer file for the ethnology collection consisted of seven master files: Africa, Asia, Malaysia, Oceania, North America, Caribbean/Central and South America, and Europe. Although the processing lab organized the files, entered the data from the inventory, and proofed the catalogue, the curators were responsible for reviewing tribal and locality names. Because some staff felt that the transfer and computerization of the catalogue cards had innumerable problems, they decided, when they were working to complete the archaeological catalogue, to enter the data not from the typed catalogue card files but from the “master ledgers” instead. Throughout the late 1970s and the 1980s, cataloguers reconfigured all terminologies, dictionaries, and naming conventions with increased care. Many meetings and memoranda from this time period document increasing awareness of the necessity of having a system that functioned coherently and that

168

O b j e c t, S p e c i m e n , D ata

properly captured the information from the original catalogue cards in a way that “made sense” within the newer computerized format. Despite this awareness, legacy data are visible in the EMu database in specific fields. These data are a legacy of the different changes or shifts in information management as the older catalogues were mapped into new technological forms. When information migrated from other media, cer­ tain terms seemed to stick to objects and the records in unexpected ways, and these could become bad, or dirty, data. In interviews, collections staff used these terms to describe data that are the opposite of “clean,” in the sense that the information has been verified by the data manager and conforms to certain rules or conventions for entering data into the system. Today, the majority of the work of collections management with respect to the database is to complete digitization projects (partly due to funding availability), and the cleanup of the dirty data that exist in records has slowed. Standardization was one way in which staff worked to help clean the data, but data can be “bad” in other ways as well. Data can have alternating and incoherent pluralizations, whereby some index terms are plural and some are not, making them unworkable. Bad data can also refer to foreign or offensive terms, and adjectives in culture names. As the data manager at the institution puts it: I’ve cleaned out some of the more obvious bad ones from other parts of the world (e.g. making sure everything is singular not plural, weeding out foreign terms and offensive terms, weeding out adjectives and culture names, etc.). But nobody has looked at the whole thing and pared it down to The One True Authoritative List of terms, so there’s still a lot that could be done, obviously. There are several overlapping terms, and trying to resolve these can be difficult. For example, who’s to say what the difference is between a “figure,” a “figurine,” and a “sculpture” ...? And then there are a lot of objects with vague index terms (“object” and “tool”) because they are unidentified objects, and there’s not much we can do about that.26

“Bad data” also includes offensive terminology or out-of-date terms for describing. In many cases, staff changed such terminology, especially terms considered to be racist or antiquated terms for tribal designations and communities. The department was particularly concerned about the use of offensive terms once the catalogue records were made available in an online portal.

C o m p u t e r i z at i o n a n d t h e L e g a c y o f D i rt y D ata

169

One staff member told me that offensive language has been a difficult issue to navigate because such terminology often establishes a historical record and trace that the museum is dedicated to maintaining. Thus, even though the terminologies have changed with respect to the active database, the pub­lic can still view the original terms used on the catalogue cards: “I started running searches for words I knew would push people’s buttons ... They’re on the catalogue cards ... not searchable in the electronic version ... but they’re on the cards.”27 Old tags with offensive terms such as “squaw” are scattered throughout the records of the anthropology collections at the museum, and their presence can negatively affect a community member’s visit to the collection. This staff member recounts a time this happened: They’re old tags ... People don’t understand if they’ve never been in a museum before or only occasionally been in a museum ... They don’t know how to interpret what they look at ... I had one of our interns ... I had her go through, just take a list of locations for one particular collection where this is a known issue. Look for the old tags and not obliterate it [the word “squaw”] but just put a pencil line through and write “woman” or “wife” or whatever the current word is ... to acknowledge that we know that people don’t like it. But in the active database it’s not searchable. So we just replaced it with either woman or wife ... It’s clear, it’s in the little brackets ... we’ve deleted something, we’ve marked it. You know that it’s a change.28

The question of how to approach changing outmoded and offensive ter­ minology is relevant to all museums that have historical records; indeed, as a Smithsonian staff member reports, rewriting this history is a curatorial decision. This individual cited the importance of maintaining a trail of in­ formation, which includes maintaining the historical accuracy of offensive terms. As an example, there was a particular group of records from South Africa that included offensive terminologies: I had a curator from South Africa ... and we had to deal with the “K” word ... at a certain point of time, it just was used as a culture term. She [the curator] said, “Unfortunately, if it says on the cards ‘Zulu Kaffir’ ... that has a meaning of [being] a particular part of South Africa, [where] those people are from; when you wipe that word out and just have it ‘Zulu,’ you lose that information, which means you have to go back and look at all the catalogue cards to find it again, because you’ve gotten rid of it ... How am I going to explain

170

O b j e c t, S p e c i m e n , D ata

that in the record without using the word?” ... I just basically said that, you know, ... [curator’s name] ... says that “the old term you can see on the catalogue card ... means that it’s from this place or this area.” It was very complicated. But that really is just such an offensive term, that if South African people saw it in the database in active use ... it would be bad.29

Later, in the electronic catalogue database, the term was changed to “Black South African” by the data manager. The balance between public engagement and maintaining a historical record on the one hand, and offending particular people, on the other, is difficult. Staff at the museum find it difficult to navigate between the ethical issues at stake and the importance of maintaining historical context. In a very real sense, there is no “best of both worlds” solution; there is only proper education about why the museum maintains old terms and why they might be useful, along with the caveats associated with this information. The Department of Anthropology generated a list entitled “Culture Terms Not in Use” in 2012 to delineate the cultural (and other) terms that were not to be used in the EMu database because of their level of offence or because they were perceived as “out of general use or confusing.”30 More “obvious” American racist cultural terms were rare in the records, but were not included in this list.31 The list therefore documents the less obvious and perhaps more controversial or confusing terms that were once seen as commonly acceptable. For example, the cultural terms document categorized Aboriginal or Indigenous terms, particularly with reference to the people of Australia, as “in flux and evolving.”32 The curator of the collections, in consultation with Australian ethnologists, was to decide on any changes to the terms. Current terminologies tend to be much more specific than the information in historical records, and so certain cultural designations can come to stand in for huge geographical areas. The example of the term “Siwish” or “Sai-wash” is particularly interesting, as it shows the level of complexity when dealing with replacing terms with newer, non-offensive ones. This is the full description of the term used in the EMu database: Siwish (SAI-wash) properly a First Nations man, but sometimes used for women as well. Nowadays considered extremely derogatory but still in use, typically with the connotation of “drunken no-good Indian.” Historically it did not necessarily have this connotation as was the generic term for Native

C o m p u t e r i z at i o n a n d t h e L e g a c y o f D i rt y D ata

171

to the point where some writers thought there was a “Siwash tribe” in the region. The origin of the word is from the French savage. When pronounced Sa-WASH, with the rhythm of the original French, it is used by modern speakers of the Chinook Jargon in Grand Ronde, Oregon with the context of meaning a Native American or as an adjective connoting connection to same (the SAI-wash pronunciation is considered offensive in Grand Ronde).33

There are significant and complex issues when navigating the Eurocentric, ethnocentric, and racist terminologies that are part of doing collections work today. For many peoples, encountering these terms online or in a database, or directly written on the tags of the object itself, can be a harrowing and confrontational experience. Changing these terms and providing correct context for them can ultimately work to heal past instances of colonial encounters. Still, as I have tried to show here, maintaining the historical authenticity of the record has a certain type of value. The issue of such terminology provides a clear example of the tension that lies at the heart of the pragmatic approach to collections care and of how collections staff encounter the normally invisible colonial infrastructure.

100-Year-Old Mistakes: Repatriation Legislation and the Importance of Documentation The most important change to museum practice in the anthropology department happened shortly after the National Museum of the American In­dian Act (NMAIA) was passed in 1989. The law first established the cre­ ation of the National Museum of the American Indian as a memorial to Native cultures and their traditions.34 It also established repatriation requirements that obliged the Smithsonian, and all federally funded institutions, to inventory collections, consult with tribes, and, when requested by culturally affiliated tribes, repatriate human remains and funerary objects. The Native American Graves Protection and Repatriation Act, which was passed by Congress in 1990, added the categories of “sacred objects,” “objects of cultural patrimony,” and “unassociated funerary objects” and applied to all institutions that receive federal funding. The NMAIA outlines specific requirements that objects must meet to be deemed “repatriate-able.” As denoted in the 1996 NMAIA amendment, to be eligible for repatriation, objects must meet three requirements:

172

O b j e c t, S p e c i m e n , D ata

1) The object must be found to be culturally affiliated with the claimant group; 2) the object must fit the definition of “object of cultural patrimony,” “sacred object,” or “unassociated funerary object” as defined in the amendment to the NMAI Act and in the Native American Graves Protection and Repatriation Act (NAGPRA (25 U.S.C. 3002(3)); and 3) the claimant must present evidence which would support a finding that the Smithsonian Institution does not have a right of possession to the object and which cannot be overcome by evidence proving that the Smithsonian Institution has a right of possession (20 U.S.C 80q-9a).35

The NMAIA and NAGPRA together required all American museums that contained objects and human remains to develop lists of the material that could be attributed to individual federally recognized tribes, so that the tribes themselves could see what was available in museum collections and decide if there were objects or ancestors they wished to be returned to them. These laws required the full cooperation of museums in finding all the relevant objects in their collection.36 In response to this legislation, the NMNH founded a Repatriation Office in the Department of Anthropology in 1991. As the NMNH website notes: This law [the NMAIA] and its amendment assert the right of Native American, Alaska Native and Native Hawaiian peoples to determine the disposition of culturally affiliated human remains, funerary objects, sacred objects, and objects of cultural patrimony currently in the collections of the Smithsonian Institution. Our policy also directs us to consider requests for the repatriation of the remains of Native American individuals whose identity is known, and objects acquired illegally. Repatriation at the National Museum of Natural History is intended to be a collaborative process, by which both staff and Native peoples work together to determine the future of human remains and cultural objects.37

According to law, only the groups of objects specified in the NMAIA (enumerated in the preceding quote) are considered to be eligible for repatriation requests. The documentation associated with objects is often the foundation upon which claims are assessed.38 The historical legacy of the early cataloguing practices is thus important to the institution in its efforts to demonstrate

C o m p u t e r i z at i o n a n d t h e L e g a c y o f D i rt y D ata

173

provenance and affiliation with a good degree of certainty. One of the principal reasons for beginning to standardize nomenclatures and index terms in particular (and not, for example, materials data) was the influence of repatriation legislation on the construction of data norms and nomenclatures at the NMNH. Part of the broader result of NAGPRA and NMAIA, as well as of the growing interest in museum collections from Native tribes across the country, was the standardization of geographic areas and “tribal nomenclature” in North America. The need for access to collections created an impetus for many projects involving the standardization of terms. Revisions to tribal nomenclatures and index terms had started much earlier, when the first computerized catalogue was compiled, but, in the mid-1990s, the Smith­ sonian made attempts to standardize these terminologies in order to facilitate complete searches for groups of objects that were from the same locality but might be in different collections or accessions. Because the NMAIA required that the Smithsonian have accurate lists of all objects from federally recognized tribes, the institution prioritized identifying misnomers and mis­ identifications in the catalogue and the database. The ability to create accurate lists of data to send to the communities was directly related to the ability to properly and accurately query the database and compile what was in the collection. Yet, this ability was compromised by the challenges I detailed earlier in this chapter. These repatriation lists had a significant impact on any individual community’s ability to reclaim its cultural heritage in the early 1990s. The ability of a community to reclaim its heritage, too, was limited in the sense that the only information the museum had was provided by the available catalogue data, which was crafted out of much older methods of documentation. Until the 1960s, the Department of Anthropology’s method of naming objects had remained relatively stable, and, as we have seen, the card catalogues were the primary way to sort (and store) collections information. By the end of that decade, computerization had necessitated deeper thought about the application of a standard nomenclature for ethnographic collections. And, with the introduction of the NAGPRA and NMAIA, the department dedicated itself to quickening the rate of the standardization of terms and material culture documentation. Indeed, throughout the process of fixing old names and data inconsistencies made durable by those who catalogued the collection, many “repatriation-related data issues” surfaced,

174

O b j e c t, S p e c i m e n , D ata

with museum staff describing a host of inconsistencies in the museum records.39 The importance of attributing primary source identification came to the fore, and the standardization of geographical terms was revised. Staff in the Department of Anthropology created tribal nomenclature standardization suggestions, and the index terms, the key finding terms in the data­ base, were to be standardized.40 Moreover, the ethnology catalogue cards were to be digitized because they contained all of the historical information that was, in many cases, unable to be properly represented in the database system. One of the outcomes of computerization was that, depending on how information was organized within the database, objects could become entirely lost if not catalogued or classified properly. If throwing sticks, for example, had not been classed as weaponry, or the wooden dance rattle had not been classed using the index term “rattle,” these objects would never be found via the computer sorts and the database searches using those terms, and they would become invisible.41 The department’s efforts to standardize tribal names, geographical location, and the index terms allowed users to search the database and obtain more accurate, or at least broader, searches. Even though it was less explicit than earlier overt attempts at classifying all ethnological heritage in terms of a functionalist and typological approach, the standardization of tribal names and index terms imposed a kind of classification scheme upon the objects. The only way access to see which objects were tribally affiliated was to assign tribal designations to them; this was done through the “Cultural Affiliation” data fields. In the second database system, INQUIRE, there were several different levels at which one could list the “culture” from which the object came. CULTURE2 became known as the primary field where cultural affiliation was recorded, and more specific information was listed in CULTURE3. The staff ’s choice to use these fields was described as purely pragmatic, and CULTURE3 was seen as a less standardized field. These fields allowed for more than one designation to be entered for an object. In INQUIRE, common challenges with searching for objects by tribal affiliation often resulted from how an object may have moved or been transferred through different communities and tribes, or from a shifting of the locality of tribes through time, so that certain tribes were present at a certain location only at a certain time. Such physical shifting meant that the catalogue sometimes contained contradictory tribal affiliations, and, in many cases, cataloguers were uncertain to which tribe an object

C o m p u t e r i z at i o n a n d t h e L e g a c y o f D i rt y D ata

175

belonged. For this reason, INQUIRE allowed for multiple terms within the CULTURE2 field, to include the possible communities from which an object had originated. Subgroups were commonly entered in the CULTURE3 field. At the same time, the spellings and named tribes were standardized, and the categorization of Indigenous peoples was solidified in the database. This was not a perfect system. “Tlingit” originally encompassed twelve subgroups: Auk, Chilkat, Hannegan, Hoonah, Hoonja, Hutsnuwu, Kake, Sitka, Stikine, Taku, Tongass, and Yakutat. These are geographical locations that, under the American administration of Alaska, came to represent different peoples because they were traditional territories of different groups. This list is an example of how museums conflated the naming of territories and people and superimposed them on pre-existing groups and cultural affiliations. In INQUIRE, staff also used the question mark an as an indicator of uncertainty; it was used to note that the data were uncertain “beyond what is indicated in normal cataloging.”42 When computerizing, applying “correct” tribal designations to each object in the collection was therefore difficult. The Handbook of North American Indians (HNAI) series documented, and in many ways regulated, the definition of American Indigenous tribal groups in the mid-1970s. Indeed, as I discussed in Chapter 4, the HNAI became the standard evidence used in the NMNH as a kind of authority file when cataloguing objects. After the NMAIA was enacted, the ethnonyms used by the Smithsonian were standardized and the spellings and political/ geographical regions were compiled from those described in the handbook. As Candace Greene put it, “Terminology used will be consistent with that of the Handbook of North American Indians unless deviation is expressly authorized.”43 Staff often saw these classifications and sub-classifications as purely practical. The database simply would not work unless there was a master inventory of the kinds of information that it could hold – in this case, index terms and tribal or geographical nomenclatures. As Greene continues, this inventory was an exercise in order to get the database to do what they needed it to do: to maintain levels of specificity that could not be handled in the current database form. [The database] is purely a practical tool, not an intellectually based classifi­ catory system. It is limited in that it cannot go beyond the catalog data that is available. If something is cataloged as “Sioux” the database cannot add

176

O b j e c t, S p e c i m e n , D ata

finer distinctions among Dakota, Lakota, and other Siouan groups. For the same reason, it cannot effectively incorporate terms currently preferred by various communities, as these terms may not completely correspond to people designated by nineteenth century usage, which may have been either looser or more specifically applied. Some terms refer to a geographic area rather than an ethnonym.44

In the context of the database, standardization was a tool for the retrieval of information, not a way to organize, once and for all, every tribe into its larger and more inclusive groups or to efface the individual differences between tribes. After the work to standardize ethnonyms and index terms was mostly completed, collections documentation maintained a reasonably stable form: the policies and procedures developed to deal with racist terms in the historical documentation, spelling errors, and shifting tribal designations all remain in use today by the collections management staff. In 2004, the new database system, EMu, was modified for the institution, and earlier data structures were migrated into the new system. Although the system allowed for more robust searches and data capture than the previous databases, there was still a need to ensure that terms and naming conventions were standardized. While policies and work procedures from this time document a desire to standardize other fields in the database, not just the index terms and tribal nomenclatures,45 the general focus was on improvements to the existing records, correcting the typographic errors and misnomers created in the nineteenth and twentieth centuries. As object records were computerized, ad hoc nomenclature rules were needed. These rules emphasized that factual statements should be short and should compel the cataloguer to rely on the physical features of the object. In this way, interpretive or intuitive remarks were left out, or perhaps left to a curator. Interestingly, the database included a space for the “authority” name – that is, the source that provided the data, whether that be a collector or a published report. Such descriptive rules assume a kind of objectivity in a consistency of practice: that the individual working with the documentation of the specimen can describe something in a “factual” way assumes that there is such a thing as a factual, unbiased opinion when it comes to the description of objects. For example, Saul Riesenberg suggests that, with the new accessioning procedures, the description of a “rattle” change as follows:

C o m p u t e r i z at i o n a n d t h e L e g a c y o f D i rt y D ata

177

Rattle On a thin hardwood stick with a rubber stopper on one end are strung, through several central holes, 33 gourd disks of variable diameter. Suggested revisions Rattle 33 gourd discs of variable diameter, drilled through centers, strung on a thin hardwood stick; rubber stopper at end.46

Here, Riesenberg recommends revisiting the older narrative description by shifting the documentation to a “technical” description. This approach emphasized the functional attributes of the object and offered no narrative description, structuring a grammar of the object that shifts the subject of the sentence to the “gourd discs” instead of the “thin hardwood stick.” These instructions also clarify that tribal names and names of villages are to be included, and any additional information cited from “authorities” is to be given in quotation marks with reference to the name of the individual who provided the information. This is a significant shift in collections documentation because of the emphasis on associating a named individual with the information provided. Often the source of the information to be cited in the description was someone other than the curator or the individual actually entering the data. In the early twentieth century, the original documentation – the accessions records, the ledger books, and the catalogue cards – were used as the primary sources of evidence for information about these objects. Throughout the development of computerization, nomenclatures were developed out of these older documents and were used with varying success, as any nomenclature was understood to be “limited in that it cannot go beyond the catalogue data that is available.”47 Regardless of the sentiments expressed by those who worked closely with the collection, the clas­sifica­ tions and nomenclatures have real effects on the knowledge organization and information management of the collection. But where do mistakes in classification and terminology originate? And why? And what have been their ramifications? Today, repatriation in the NMNH often originates with an Indigenous community whose members have become aware of a specific object in the museum’s collection. The fulfilling of a repatriation request requires detailed investigative research on behalf of both the community and the institution.

178

O b j e c t, S p e c i m e n , D ata

At the NMNH, the Repatriation Office handles these requests. Steps in ful­ filling the requests include the museum discovering if the collection holds any objects of cultural affiliation; documenting whether or not the objects are still physically present in the collection; determining if they fit a definition under the law; and, finally, proving that the museum did not have the rights to obtain the objects in the first place and that the ownership lies with the community. If the objects in question do not satisfy the requirements for any one step, then the next step is not evaluated. Therefore the historical information and the material technologies used to record this information have vital implications. In some cases, the way that objects were originally documented impacts their ability to be considered for repatriation requests. If there are mistakes in the documentation, the museum often still holds authoritative power over the process, although this is changing. Repatriation provides an example of the enduring legacy of routine work practices from a century and a half earlier, and reveals the ethical dimensions and challenges raised when an information infrastructure becomes materially durable. Although part of normalized museum practice at the Smithsonian for over twenty years, repatriation raises significant issues about the construction of the catalogue and its categories of description. Making the collections accessible is the first step in much repatriation work, but putting collections online has presented its own set of opportunities and challenges.

Making It Public: Issues of Online Access In the first decade of this century, as part of a broader institutional effort to increase access to the museum’s collections, the Department of Anthro­ pology decided to make the catalogue information from its EMu database available online. This decision required the department to review its data, as it was concerned about providing open access to potentially problematic information. Staff discussed what kinds of data fields would be available to the public, and they reviewed and reworked some of the data, particularly on “sensitive” items, item locations, and offensive terminologies, to ensure that collections would appear in the way that the department wanted them to.48 In 2006, the department decided that notes and accession data would not appear for any records at all, and that old, outdated, or offensive terminologies would be kept in EMu but would not appear on the web version. Later that policy changed, and the notes did end up appearing for all records.

C o m p u t e r i z at i o n a n d t h e L e g a c y o f D i rt y D ata

179

In preparation, the department conducted a project to extract any locality data from notes in archaeology records; these data were kept offline. The department also decided that the images of ethnology catalogue cards, but not the archaeology ones, would be available online: archaeological collections information could hold very specific locality data, which could expose the location of archaeological sites to potential looters, and it is common museum practice to withhold such data. This level of access to the collections increased awareness that many of the attempts at standardization, including the creation of index terms, were a part of institutional practice but did not necessarily result in intuitive ways to search the collections. Standardized terms were created specifically for researchers and in-house collections staff. The cultural area designations, tribal nomenclatures, and the index terms all reflect this purpose. Ultimately, the online version of the collections database included all ethnology records, almost all archaeology records, the full notes field for all of the records, and most locality information, except for specific archaeological sites, human remains, or inactive records like missing, exchanged, or de-accessioned ob­ jects. The public fields for the database were: Division, Object name, Index term, Other numbers, Culture ethnicity, Locality (continent, country, state), Site name, Site number (but not coordinates), Collectors, Donor, Date collected, Accession number, Date accessioned, and Multimedia. In addition to the public fields, the staff-only displays incorporated most, if not all, of the available fields within the EMu system: Specimen count, Materials, Techniques, Accession notes, Date created, Notes, Storage location, Precise locality, Creator, Associated events, Associated individuals, Object description, Bibliographic references, Culture 1, Current and old cultures, and Current and old object name/index term. Some other data restrictions were followed in the online version of the database. Physical anthropology data (skeletal and non-skeletal) were excluded. Second, certain catalogue records themselves were marked to prevent specific images from being available to the public, including those that raised issues of “cultural sensitivity, copyright, preservation of archaeological sites or other legal issues.” In addition, fields from the physical anthropology database were left out of the public website because of the lack of data present in them and because some data were “questionable at best.”49 Similar issues arose regarding public access to the archaeological collections. In particular, the department had concerns about including the locality

180

O b j e c t, S p e c i m e n , D ata

fields for fear that overly specific information might contribute to the destruction of a site.50 The report of an advisory committee explains: While it was acknowledged that ... more general information, such as “3 miles south of ” could potentially be used to help locate sites, it was agreed that on balance the benefits of making information available to researchers outweighed the risk of misuse. Discussion: our recommendations regarding inclusion of records and fields are based on a strong desire to make our online database maximally useful to researchers, while exercising reasonable prudence in disclosure of sensitive data.51

Given the fact that these locality data were embedded across three fields (locality, notes, and catalogue cards), the department decided that the “finer” locality levels (such as longitude and latitude) not be accessible in the public view. However, it was eventually decided that, because the notes field includes some of the most important information in the database, it should be accessible. Staff identified specific problematic records and amended them by moving location data from the notes field into other fields that would not be displayed on the website. This involved the review of over 13,000 suppressed records. By the mid-1990s, almost all the card catalogues for ethnology had been scanned, and they are available as part of the online record, in part because it became more efficient to show the cards rather than to replicate them in the system.

How the Database Was Done As I have demonstrated in earlier chapters, the transition from paper catalogue cards to the computerized system was not just a simple copy-and-paste solution; it required intellectual work to modify the records and the names, which would then be solidified in the computer system. Such selection and solidification create “legacy data,” which persist through time and are made visible in unexpected ways in the current database. Such data have raised complex ethical issues about the suitability of historical terms for contemporary museological work. Museum cataloguing work is infrastructural work, whereby the “daily work of one is the infrastructure of another.”52 Once embedded into the information infrastructure, the routine work practices

C o m p u t e r i z at i o n a n d t h e L e g a c y o f D i rt y D ata

181

become invisible. In this way, the contemporary database, imbued with these historical legacies, is a performance of earlier colonial-era collecting and is a process rather than a stable documentary form. Indeed, as the NMNH’s online collections database notes, “these data represent a work in progress.”53 “Dirt,” as Mary Douglas first put forth, is “matter out of place.”54 In this chapter, I have shown that data too can be dirty and out of place in time. Due to the history of naming and exclusion, the legacy of the data in the records at the Smithsonian is that such data became dirty – and they need to be cleaned. This is a different kind of “dirtiness” than in the early years of the institution. Where “good” data were once associated with the value of information for the progress of science, now good, clean data are associated with the desire to not harm, or even to undo harm. Yet, a kind of in­ visible work is still perpetuated by reiteration and reinscription, part of an infrastructure that becomes visible only in the breakdown. Categories and ways of doing become inescapable if left unaccounted for. Many mistakes had been made when the catalogue cards were transcribed into SELGEM, INQUIRE, and EMu, and attempts to fix these errors were not completely successful. Ultimately, the department decided that the catalogue cards were to be digitized and attached to the records in order to preserve the information on them. Although the practice of data migration was part of routine work – a kind of daily practice – it draws attention to the complex nature of information infrastructures, and it highlights the local and material practices – and unrecognized labour – that exist in these spaces. These records were taken “as is” for authority files and were made durable by their re­ inscription into newer computing software – in this process, they can become the dirty data. Collections documentation during the mid-1960s and 1970s had stabilized the earlier object descriptions and naming conventions through translation and mediation into new “digital” technologies. As John Law reminds us, some materials are more “durable” than others, and they maintain a relational pattern for longer.55 This concept is useful when considering the catalogue and its digital representation, and for understanding the sticki­ ness of certain terms and modes of thought through time, and the implications the stickiness has for public access. The ledgers and catalogue cards, despite their potential misattributions and lack of information, were inscribed in the computerized system as a way to preserve the information about objects in the collection. As a consequence, the repatriation of objects

182

O b j e c t, S p e c i m e n , D ata

is governed by categorical distinctions that are deeply historical and that do not necessarily represent an accurate vision of reality or of the importance of things. In current legal and ethical frameworks, the assumed authority lies with the institution or the federal government, and I hope to have shown here how this authority can be, and is, questioned by those who work to repair these legacies.

C o m p u t e r i z at i o n a n d t h e L e g a c y o f D i rt y D ata

183

Conclusion

A Museum Data Legacy for the Future

Molten in their form, colonial entailments may lose their visible and identifiable presence in the vocabulary, conceptual grammar, and idioms of current concerns. It is the effort of this venture [i.e., her study] to halt in the face of these processes of occlusion and submersion, to ask about how they work, their differential effects; and on whom they most palpably act. – Ann Stoler, Duress

Tribal nations rarely have full access to or knowledge of documentation, especially when related objects and collections are housed in different repositories. The NAGPRA and NMAIA object categories – “sacred” and/or “patrimonial” and/or “funerary” – are often determined not just through consultation but through the chance preservation of contextual evidence. What rituals were enacted when those objects were in use? Who permitted that excavation? Where is the proof of tribal identity? Were those things personal property or national patrimony? Who decided then? Who decides now? When Native nations and museums disagree, museums retain the final say. With all this in mind, it seems fair to ask if federal attempts to facilitate repatriation have actually restored respect for Indigenous patrimony writ large, or merely legislated respect for a few protected (and museologically accepted) categories of Indigenous materials. – Margaret Bruchac, Savage Kin

Do the histories of museums and their documentation technologies foregrounded in this book preclude an ability to change? If, as Elizabeth Povinelli has warned, the “postcolonial archive will never be compatible with the colonial archive,” then is a postcolonial archive or museum catalogue a possibility?1 Do we need, as Hays-Gilpin and Lomatewama suggest, other ways of understanding objects and belongings that are specific to each community they come from, where they may even be beings in and of themselves?2 These questions are not necessarily new, and they are centrally about representation, difference, and power.3 This book has been an effort to under­ stand the ingrained infrastructural qualities of record keeping, in an attempt to do justice to heterogeneous assemblages of people and material technologies that manufacture knowledge in museums. I do not present an easy explanation of why colonial formations and epistemologies gained prominence, or why Indigenous knowledges were obscured, or why colonialism had a significant impact on modern national cultural heritage institutions. Rather, I have been more curious about how the structural erasure of knowledge can take place, how it was made durable through practice and daily work, and the nature of invisible labour that looks to repair these issues. Recent criticism has again brought the violence of colonial collecting to the fore, and what museums worldwide need to do to address this violence.4 If we are to change practice broadly, we must understand how colonialism is done on a daily basis, and how it is ingrained in material technologies. In the words of Ann Stoler, we must describe how colonial entailments lose their identifiable form, but still retain a molten, shifting presence. I have been especially concerned with how the values around object data and human data have changed throughout time, but how they also paradoxically remain durable in material technologies. For example, those who have the (informal or not) rights to award claims of veracity to information have rights to the legacies of that information as well. This (as Margaret Bruchac points out) is of key concern. Through an examination of the material durability of collections documentation at the National Museum of Natural History, I have questioned different modes of organization and critiqued the epistemological underpinnings of contemporary museological (and, broadly, knowledge) work.5 By no means do I intend my critique as an insult to the important work of current cataloguers and museum staff, who know these issues far better than I; and it is true that there are structural museum issues worldwide that take precedence over and divert funding away

A M u s e u m D ata L e g a c y f o r t h e F u t u r e

185

from cataloguing. Understanding the history of record keeping is an important part of decolonizing national institutions in the future, and a process that could include dismantling perceived authority and control. This analysis could cause us to question other, more pervasive ways through which power and authority are embedded within documentation. The production of ethnographic objects is thus simultaneously the production of knowledge and identity: of the Smithsonian as a scientific research institute, of the actors and their daily work, and of the communities whose cultural heritage is still framed by the museum. Ethnographic objects are made into specimens and maintained in this web of meaning through the information infrastructures that mediate their use. To study a history of cataloguing and documentation is to problematize the category of the “ordinary” and examine how it is a “delicate construction, made and re-made every day.”6 In the case of the Department of Anthro­ pology at the NMNH, the cataloguing practices were normalized through work and daily routine, and stabilized as infrastructures embedded into documentation. They are likewise continually practised and remade through constant performance and inscription. The practical politics of the museum are made manifest not only in exhibits and displays, but also in the material durability and performance of terms, naming conventions, and the historical fields applied to the documentation of cultural heritage. The temporal “build-up” or “stickiness” of terms, categories, and practices, which is linked to earlier epistemological modes, constitutes a socio-technical information infrastructure. The museum catalogue – digitized or on paper – can be considered an object with its own history. In reading the material documentation in formal and informal archives, and through speaking with current museum staff, I found this history to be complex. My account of it begins with an analysis of the epistemic commitment to studying the natural historical sciences at the NMNH in the nineteenth century, and details how modes of objectivity were present in the study of material culture. Documenting objects became an important practice at the institution, used to ensure rigorous standards and to legitimize ethnology as a “good” science. In the late nineteenth century, there was a shift when a particular individual, Otis Mason, became obsessed with applying the principles of classificatory order to the collections. The ideologies and epistemologies of this era translated directly onto object records and into ledger books through the naming of objects and by including some kinds of information and excluding others. Through the

186

C o n c lu s i o n

collection of ethnographic specimens, and the many attempts to establish classificatory concepts, certain characteristics were made to “stand in” for the object as a whole. These are represented in the fields that were developed to contain the object’s information in the ledgers and later in the catalogue cards. Recording objects made them specimens, or objects of scientific knowledge, and this preceded their exhibition – which subjected them to the public and the exhibitionary “gaze.” The naming strategies and documentation of the “good specimens” that were collected, labelled, and recorded persist today, and are important historical traces that are figuratively and literally attached to objects. Mason’s influence over the development of the catalogue was great. It was his dream of organizing and creating an encyclopaedic catalogue that launched the museum into a new era of record keeping with the adoption of the card catalogue. As I document in Chapter 3, the end of the nineteenth century saw an increase in the tension between internal consistency and standardization – and of change or iterative development. Despite the desire to scientifically classify and name objects, such processes were often haphazard. Yet the documentation still used earlier visual rhetoric that was strongly situated in ideals of scientific natural history collecting. Card catalogues are conceptual precursors to computerization, and, as Markus Krajewski has argued, workers and bureaucrats saw moveable index cards as machines.7 Their adoption by the NMNH (and likely elsewhere) mirrors an increasingly mechanized and routinized process of object documentation. The museum paid increased attention to the development of work practices and a systematized approach to cataloguing, and the card catalogue was maintained as the main system of object documentation until the mid1960s, when the Smithsonian adopted computerized database systems to house information. The twentieth century saw broad changes in the discipline of anthropology, yet many of the practices formalized much earlier remained relatively stable. Computerization eventually raised new issues that the museum did not necessarily present as “intellectual” or “epistemological” but, rather, as practical. These issues were tied to the development of the new technologies themselves. As I explored in Chapters 4 and 5, the standardization of terms in the anthropology catalogue was required because there had been no consistent naming system in the collection for over 150 years. The computerized systems needed to be coded or “told” which items were related and which were not. The staff at the Smithsonian spent years trying to establish

A M u s e u m D ata L e g a c y f o r t h e F u t u r e

187

an object-naming system that allowed them to search by object type, ethnonym, and materials. The negotiation of relations between structures and agency,”8 or the tensions between local or specific and generalized standards, became the central task of the Department of Anthropology during this time. Ultimately, computerization accomplished several things: it contributed to a standardization of terms, it made possible greater access to collections information, and it further routinized work processes with a commitment to organizing material culture from a pragmatic perspective. Even as epistemologies changed in the twentieth century, the repetitive nature of the work, the devaluation of this labour, the adoption of normative practices, and the application of new recording technologies, such as the card catalogue and the computer, enabled a stability in the way that cultural heritage was named and categorized. Today, the normative practices that assumed a collector’s authority and fostered an absence of Indigenous words and makers associated with an object become visible only when the system does not work: when it becomes impossible to find objects by their name in the database; when repatriation claims necessitate an excavation of the traces of past classifications; or when people themselves are faced with continued micro-colonial acts as they encounter legacy data inscribed upon objects or in collections records. In this book, I have made visible the processes and practices that construct the normative idea of ethnographic objects in museums, and how this can reinforce and re-inscribe value onto data. The idea that data have colonial legacies raises an important issue, not just for museums and archives but for anyone who uses technological systems to describe objects and people. Understanding how legacy data are inscribed in collections records is an ethical issue that is of consequence in an institutional climate where historical names are often maintained and encountered by communities who work with these collections on a regular basis.9 Legacy data constitute an example that demonstrates how the information attached to the object record through multiple data migrations is manifest in current work. I have used the term throughout for what it means, but also for what it connotes – that is, that data have histories, and these must be examined. The issue is not simply that objects may have been collected within a particular Eurocentric worldview; they were also documented, classified, and recorded as such. Owing to the bureaucratic practices of transcription and mediation, the categories and naming strategies “stuck.” The durability of these terms is afforded by material technologies; and, as this book has shown, the terms are inscribed and reinscribed upon new sets of

188

C o n c lu s i o n

technologies through time. The persistence of racist terminologies, mistakes, lost objects, and outdated or incorrect object names speaks to how object documentation norms “stick” to both technologies and practices. These per­ sistent traces are not trivial. Infrastructures organize knowledge, and they have material force. Ways of delineating hierarchy and order are built into and embedded in every feature of the environment and “are a mixture of physical entities such as paper forms, plugs, or software instructions encoded in silicon and conventional arrangements.”10 Understood as an “infrastructure,” the Smithsonian’s catalogue is material, or materially textured, and it is this key principle that makes it durable. The construction of ethnographic cultural heritage is built upon paper and digital technologies as well as human beings who make and modify the records. Information recorded in catalogues and databases acts as a form of long-lived material or physical institutional memory – or, imagined differently, as a form of forgetting. Histories of acquisition, the extent of staff members’ knowledge or time, and the joining and dissolution of museum departments all leave their marks in the information recorded in a database. All of these constitute the invisible “work” of the infrastructure.11 The history of past practice is written and translated into new material technologies, no matter the epistemological shifts that occur. The museum catalogue is thus the internal mechanism of the museum, one that is increasingly masquerading as “unmediated” online access to the “raw data” of the collection. In my recent conversations with Dr. Candace Greene, the previous collections manager at the NMNH, she spoke about the changes in cataloguing through time, and the importance of understanding both the possibilities and limitations of databases. Principally, she argues, museum work is about tangible things – about objects and their importance in history and to communities of origin. Her career has spanned museums and lasted well over forty years, and, in her experience, cataloguing has often been completely free-form, and practised by people without training. Most individuals assigned to this task, both today and in the past, have been unfamiliar with specialized terminologies or cultural specificities; indeed, no one person could have this breadth of knowledge for materials from around the world. Moreover, the attitude that cataloguing is “deadwork” is still common in many museums. Cataloguing is often still handed to the least skilled, least knowledgeable, and least experienced worker – whether a volunteer or an intern. The training in this area rarely has more to offer than how to create a record, use a few authority files, and then save the record in the system.

A M u s e u m D ata L e g a c y f o r t h e F u t u r e

189

Yet engaging with objects as cultural productions is the paramount work of anthropology museums; and working with objects requires an understanding of how records are created and managed – historically and today. As I have shown, records are the starting point of almost all of this work. Today, these are computerized records. Things that are not indexed will not be seen by researchers, unless they have the time, access, and financial support to walk slowly through the stores and open cabinets one by one. Access to accessions files, ledgers, catalogue cards, and other documentation is simply invaluable in this context.12 Active collecting still happens at the NMNH, but nowhere near as much as it did in centuries past. Now, a necessity in cataloguing is to enhance old records, something Dr. Greene has dedicated her current practice to. Often, researchers brought into the Department of Anthropology and its connected archives end up generating far more information than there is staff to absorb, and Greene questions whether the data­­ base is the appropriate place to hold often nuanced cultural information. My examination of the intellectual and social history of the Department of Anthropology’s catalogue raises significant issues for future scholarship. Some of these are related to the practice of object replication at the Smith­ sonian, which was, at the time this research was conducted, new and replete with potential challenges and uncertainties. Replicating the Hoonah Repatri­ ation Collection in 3D, for example, raised concerns about and possibilities for addressing the theoretical implications of replication and digitization technologies.13 Now, museum objects are digitized and reproduced at increasing speeds, and, as Haidy Geismar has observed, 3D printed objects (for example) circulate as representations of time, possibility, and politics. She correctly asserts that reproducing objects digitally has “shifted the weight of intellectual authority away from collections into their digital counterparts.”14 Such practice raises the question, what are the histories of our contemporary media technologies, and what epistemic loyalties do they reproduce? Certainly, capturing a bust of Nefertiti (for example), reproducing its 3D image, and providing that image for free on the Internet is a political act that raises our awareness of ownership and access.15 It is also a provocation, one that urges us to question if these new digital models are representations of objects or if they are, in fact, wholly new carriers of meaning and knowledge.16 Future analyses could integrate museum anthropological studies of copies and reproductions with information studies work that concerns the concept of objects as documents and carriers of information.17 Are digital replicas

190

C o n c lu s i o n

different from earlier models? Do the digital reproductions of objects have a different, or new, value of authenticity and veracity of form? More practically, museums must address concerns regarding the documentation of digitally replicated models, such as the intellectual property of the scan or photogrammetry data. If actual objects can be replaced with 3D models, could the museum print all objects collected from Indigenous communities and return the originals?18 What other kinds of reconciliatory work needs to be done to shift the power of the museum in these cases? What new museum practices could these technologies afford? The social practices of early computerization often relied on a deference to the “purely technical,” and today we see this ideology prevail. The affordances of specific technologies often elide ethical consideration – if a technology can do something, it should. Yet, as I hope to have shown, the ways that individuals imagine themselves in relation to the technologies are central to any kind of decolonizing perspective. As new digitized representations proliferate, work must be done to consider how to navigate the complex social, political, and ethical considerations of this new type of specimen data, moving beyond a purely pragmatic approach to collections documentation to a focus on the protection and appropriate use of knowledge broadly.19 Another concern comes from my experience working at the intersection of museum data and community-based research.20 When artists, researchers, and other experts come to work with museum collections more broadly, their knowledges are often slotted in as part of a “narrative” or long-form description that has significant effects. Framing Indigenous or other community knowledges as a narrative can assume that museum fields (which have been associated with “Western” science) are more fixed, authenticated, and valid than the free-form narrative fields that often incorporate stories and personal information. The evidences and histories presented in this book foreground the fact that museum fields in the database are narratives too: they are a part of a kind of institutional oral history. In this way, they are likewise constructed, situated, local, and performative. Including Indigenous knowledge as part of a narrative field in the database but not changing the nomenclature of objects may do little to frame the objects themselves in a meaningful way for any community whose voices are not included in official or national discourses. Museum practice has begun to consider the “practical politics” of classifying and naming material culture, but more research is needed to determine how and why bureaucratic documentation like circulars, ledgers, and card catalogues were adopted more

A M u s e u m D ata L e g a c y f o r t h e F u t u r e

191

broadly by other early museums at the time and in each specific situation. Such study requires an in-depth examination of other early institutions in North America and Europe to determine how the network of objects, “data,” and actors functioned to produce broad ethnographic ideas of material culture across museums, universities, and public audiences. Although there has been much work in developing Indigenous-specific databases that are outside of a museum context,21 we must consider how – and even if – museum documentation can change in the institutional setting, or if it can fulfill other information needs in communities. This is fundamentally a question of difference: Can an institution, whether the NMNH or another institution, include entirely different ways of knowing and still create a functional system? Can practice move beyond technological pragmatism to a more ethical approach to documentation that is more nuanced? Should the catalogue be considered an appropriate place to hold valuable and intangible aspects of heritage, like stories, songs, and languages? Or are there other more appropriate venues through which Indigenous peoples themselves can control, create, and disseminate objects and data?22 And how do we return objects, information, and power? Focusing on cataloguing standards and new ethical requirements in constructing metadata and catalogues is beginning to be addressed in important ways, and not only within museums but archives and libraries as well.23 Simply put, such a focus requires increased funding for the management of collections, not just exhibits development. Making sure funders understand the importance and relevance of object documentation and basic digital preservation is therefore paramount.24 The Smithsonian and the NMNH are both highly active research sites with numerous programs that conduct critical work in repair and reconciliation. Repatriation is still part of the daily work of the museum, and its digitization projects show potential for longstanding change. Further, there are a myriad of other projects, individual researchers, and workers who contribute to the NMNH’s goal of seriously interrogating and changing these histories. For example, the Recovering Voices program funds research on linguistic revitalization in many different communities, 25 and the Sum­ mer School in Museum Anthropology (SIMA) was established as a way to train researchers and academics in the languages and complexities of doing work in museums and with records.26 Other Smithsonian museums, in particular the National Museum of the American Indian, have worked to redress missing histories and present future visions of Native art.

192

C o n c lu s i o n

Another path for more research builds upon attempts to reconstitute the history and legacy of classifications, standards, and legacies writ large, across many different kinds of institutions. Important groundwork has laid bare the necessity of thinking with and through standards and classifications in libraries, for example, and more work is needed to determine when, where, and how classification and standard systems came into use and for what reasons. Following how these systems change through time, particularly in the human sciences, requires more in-depth historical study as well as attention to the media archaeologies of these systems. In this book, I have foregrounded the historical and social origins of what we tend to see as normal, daily information practices. Modes of cataloguing cultural heritage are performed and made durable in infrastructure. When used as authoritative documents, catalogues shape the relationship between ethnographic knowledge and cultural heritage. Practices and technologies, embedded in catalogues, have ethical consequences, yet paying attention to other histories is often difficult in this context because they are so often completely occluded. But a deep, epistemological history, such as the one presented here, addresses how material culture has been framed, named, routinized, and valorized. Opening this black box of museum infrastructure helps call into question other normative procedures implicit in information systems more broadly. The epigraphs provided at the beginning of this book were of two specific bureaucratic moments: a document from an institutional archive that specified the kinds of information that were required in order to authenticate and validate objects for scientific study at the turn of the twentieth century, and an excerpt from late twentieth-century legislation mandating the return of cultural heritage to Indigenous communities in the United States. Although much time separates these two documents, both represent the legacies of settler-colonial imaginations in the history of material culture. Looking deeply at the sociotechnical practices of documentation in a museum anthropology department is one way to grasp how we can begin to make strange, to upset, and to deconstruct the normalized and moralized institutional infrastructures within which all these practices are embedded.

A M u s e u m D ata L e g a c y f o r t h e F u t u r e

193

Notes

Introduction: “The Making of Specimens Eloquent” 1 The NMNH was called the United States National Museum (USNM) from about 1881 to 1957. 2 In addition, I have published several articles on critical cataloguing: “Critical Histories of Museum Catalogues,” “The Computerization of Material Culture Catalogues,” “Decolonizing Ethnographic Documentation,” and “Organizing Knowledge in Museums.” 3 “Studying up” is a term taken from Sandra Harding’s feminist philosophy, standpoint theory, which calls for the studying of scientific disciplines from other, subaltern lenses. For more, see Harding, The Feminist Standpoint Theory Reader and Sciences from Below; Wylie, “Why Standpoint Matters.” 4 I use the term “Indigenous” throughout this book but am conscious that it is necessary to refer to specific communities by name, as there is no shared Indigenous philosophy or history that spans the globe. The broad term “Indigenous” has global political resonance today but may not be the term many identify with (see the United Nations Declaration on the Rights of Indigenous Peoples, UNDRIP). Other terms that some may identify with are Ab­original, Inuit, First Nations, Métis, First Peoples, Indian, and Native. 5 Scholars have looked at the politics of decolonizing anthropology, archaeology, museums, and curatorial practice broadly, and with a contemporary focus. A few, but by no means all, are listed here: Agrawal, “The Politics of Indigenous Knowledge”; Anderson and Christen, “Decolonizing Attribution”; Atalay, Community-Based Archaeology; Collier-Wise, “Identity Theft”; Isaac, “Decolonizing Curatorial Practice”; Lonetree, Decolonizing Museums; MoretonRobinson, The White Possessive; Nakata et al., “Decolonial Goals and Pedagogies for Indigen­ ous Studies”; Phillips, Museum Pieces; Simpson, Making Representations; Smith, Decolonizing Methodologies; Tuck and Yang, “Decolonization Is Not a Metaphor”; Vizenor, Survivance. 6 Duarte and Belarde-Lewis, “Imagining,” 677. 7 Ibid., 683–4. For more on knowledge organization from an Indigenous perspective, see Littletree and Metoyer, “Knowledge Organization.” 8 Duarte and Belarde-Lewis, “Imagining,” 681.

9 I use the term “belongings” occasionally in this book. This term has been advocated for by xʷməθkʷəy̓əm (Musqueam) scholar Jordan Wilson in his curatorial practice, and also by the work of Kate Hennessy’s Making Culture Lab. “Belongings” has been used to describe objects in museums, particularly on the Northwest Coast. The term connotes that they rightfully belong elsewhere – namely back in their communities and with their families. For more, see Wilson, “Belongings in Cə̓ snaʔəm”; Muntean et al., “‘ʔeləw̓ k̓ ʷ – Belongings”; and Muntean, Turner, and Hennessy, “Addressing Digital Affordance.” 10 Bruchac, Savage Kin, 176. 11 Ibid., 17. There are also occasional examples of where anthropologists held material out of their work for fear of harm to an individual or community. 12 Rosenberg, “Data before the Fact,” 19. 13 Ibid., 37. In this article, Rosenberg is responding to the work of Mary Poovey, who has similarly claimed that facts are historical and contextual. See Poovey, A History of the Modern Fact. 14 Historically, this idea was taken up by sociologists and historians of science. See the influential works of Karin Knorr-Cetina, The Manufacture of Knowledge; Thomas S. Kuhn, The Structure of Scientific Revolutions; and Ludwik Fleck, Genesis and Development of a Scientific Fact. See also recent work on the biobanking and crafting of specimen genomics: Van Allen, “Pinning Beetles, Biobanking Futures.” 15 For a good overview of these ideas, see Lisa Gitelman and Virginia Jackson’s introduction to “Raw Data” Is an Oxymoron. 16 Riles, Documents. See also Hull, “Documents and Bureaucracy.” Hull argues that, although documents are important in early anthropological endeavours, we cannot define anthropology simply as a bureaucratic practice. This is true, but, as I argue, documents take on a more central role in the context of museums. 17 Repatriation is both a legal requirement for American museums and also a much broader, more complex issue that spans borders and time periods. Successful cases rely on the research acumen and activism of (largely unpaid) community members and the work of museum repatriation liaisons, researchers, collections managers, and curators. It is a topic I am not covering in detail here. For more on repatriation, please see Bell and Napoleon, First Nations Cultural Heritage and Law; Bray, Reckoning with the Dead; Chari and Lavallee, Accomplish­ ing NAGPRA; Colwell, “The Sacred and the Museum”; Conaty, “The Effects of Repatriation on the Relationship between the Glenbow Museum and the Black­foot People”; Echo-Hawk, “Keepers of Culture: Repatriating Cultural Items under the Native American Graves Protection and Repatriation Act”; Forde, Hubert, and Turnbull, The Dead and Their Possessions; Krmpotich, The Force of Family; Merrill, The Return of the Ahayu:Da; Trope and Echo-Hawk, “The Native American Graves Protection and Repatriation Act”; and Webster, “From Colonization to Re­patri­ation.” 18 Here I call to philosophies of science, specifically how scientific objects come into being. For more on the concept of agential realism (that objects exist in the exchange between observer and material thing), see Barad, Meeting the Universe Halfway and “Posthumanist Performativity.” 19 To clarify, the NAGPRA did recommend that museums consult with tribal authorities in order to compile inventory lists of collections, and it did allow for other pertinent sources of data to be considered. This book is not meant as a criticism of the legal pathways the NAGPRA has opened for the return of material culture. The law does not explicitly assume that museums will become the primary source of provenance information; but, often in these cases, museum records are the only documentary traces that survive, and can become the de facto sources.

N o t e s t o pa g e s 7 – 8

195

20 Douglas, Purity and Danger, 44. 21 Ibid., 45. 22 Pratt, Imperial Eyes, 15. 23 For more on order and difference in museums, see Dias, “The Visibility of Difference”; Bennett, The Birth of the Museum; Bennett, Pasts beyond Memory. 24 Pratt, Imperial Eyes, 16. 25 Ibid., 25. For more on Linnaeus’s classificatory paperwork, see Charmantier and Müller-Wille, “Carl Linnaeus’s Botanical Paper Slips.” 26 Hetherington, “From Blindness to Blindness”; Foucault, The Order of Things. 27 Daston and Galison, Objectivity. 28 Jenkins, “Object Lessons and Ethnographic Displays.” 29 Bennett, Pasts beyond Memory, 77. 30 Pratt, Imperial Eyes, 29. 31 Foucault, The Order of Things, 137. 32 Pratt, Imperial Eyes, 25. 33 Ibid., 32. 34 The archaeological sciences adopted these theories earlier than did ethnology. John Lubbock (Lord Avebary) has been credited with the application of a Darwinian evolutionary approach to archaeology, which was used to explain the technological development of Europe in opposition to that of North American Indigenous populations. See Trigger, A History of Archaeological Thought, 114. 35 Bennett, Pasts beyond Memory. 36 Another example of the naturalization of patriarchal norms is that of taxidermy displays in the American Museum of Natural History. See Haraway, “Teddy Bear Patriarchy.” 37 Bennett, Pasts beyond Memory, 160. 38 Ibid., 65. 39 Jenkins, “Object Lessons,” 257. 40 Daston and Galison, Objectivity. 41 Ibid., 368. 42 Daston, “Type Specimens,” 156. Daston and Galison’s project makes way for a relativistic, ethical understanding of the history of science without destroying its merits and practice. As they argue, “to historicize ... analyses does not ipso facto invalidate them. It does, however, unsettle their self-evidence” (Objectivity, 378). 43 Fenton, “The Museum and Anthropological Research,” 330. 44 Ibid., 329. 45 Ibid. 46 For more on the connection between anthropology and colonialism, see Barringer and Flynn, Colonialism and the Object; Clifford and Marcus, Writing Culture; Pratt, Imperial Eyes; and Reid and Paisley, eds., Critical Perspectives on Colonialism. 47 Brown and Peers, Museums and Source Communities; Krmpotich and Peers, This Is Our Life; Knight, “Unpacking the Museum Register.” 48 Further reading on collecting includes Bell, “A Bundle of Relations”; Bennett, Pasts beyond Memory; Isaac and Isaac, “Uncovering the Demographics of Collecting”; Lubar, Inside the Lost Museum; and Pearce, On Collecting. 49 Bowker and Star, Sorting Things Out. 50 This is similar to arguments made by Lorraine Daston and Peter Galison concerning the history of objectivity in the sciences in Daston and Galison, Objectivity. 51 For example, Lisa Gitelman’s historicization of documentation in Paper Knowledge has led me to consider the role of bureaucratic practices and tools in establishing shared understandings of evidence, data, and objectivity; and Kirsten Weld, in Paper Cadavers, has recently

196

N o t e s t o pa g e s 9 – 1 4

examined a particular case of how the police archives of dictators in Guatemala, found crumbling in old buildings, came to represent the deceased or tortured bodies they depicted. For more on infrastructure as a concept of concern, see Bowker et al., “Toward Information Infrastructure Studies”; Lampland and Star, Standards and Their Stories; Mattern, Code and Clay; Parks, “‘Stuff You Can Kick’”; Star and Ruhleder, “Steps Toward an Ecology”; and Turner, “Organizing Knowledge.” 52 For more excellent work in this area, see Halpern, Beautiful Data, and Jonathan Sterne, MP3. 53 Wright, Cataloging the World; Krajewski, Paper Machines. 54 For example, see Campbell, Agitating Images; Edwards and Hart, Photographs Objects Histories; and Griffiths, Wondrous Difference. 55 Hochman, Savage Preservation. 56 Lemov, Database of Dreams. 57 Law, “Notes on the Theory of the Actor-Network,” 387. See also Latour, Science in Action; Vygotsky, Language and Thought; and Strathern, Partial Connections. 58 Law, “Actor Network Theory and Material Semiotics,” 149. 59 Ibid. 60 Ibid. 61 Stoler, Duress, 7. 62 Theories of performance have a long lineage, particularly within feminist scholarship. For a few of particular interest, see Bauchspies and de la Bellacasa, “Feminist Science and Technology Studies”; Butler, Gender Trouble; Haraway, “A Manifesto for Cyborgs”; Harding, Sciences from Below; Stoler, “Colonial Archives and the Arts of Governance”; and Stoler, Along the Archival Grain. 63 Mol, The Body Multiple. 64 Likewise John Law, building on earlier work by Bourdieu, argues that the “performative” is a key concern of material semiotics. See Bourdieu, Outline of a Theory of Practice, and Callon, “Some Elements of a Sociology of Translation.” 65 Waterton, “Experimenting with the Archive,” 650. 66 This formulation is bolstered by other theories of performativity that place the existence of objects and categories in their continued construction – their “doing.” As Waterton declares, “Performances, in other words, always end up being subtle and flexible improvisations rather than strict replications of existing realities” (ibid., 651). See also Szymanski and Whalen, Making Work Visible. 67 Haraway, “A Manifesto for Cyborgs.” 68 Parezo, “The Formation of Ethnographic Collections.” 69 The term “blanks” is used by Lisa Gitelman in Paper Knowledge. 70 Much of this relies on the work of Catherine Nichols; see “A Century of Circulation.” 71 The full name is Q?rius, the Coralyn W. Whitney Science Education Center. 72 Hollinger et al., “Tlingit-Smithsonian Collaborations.” 73 More information about this repatriation can be found in ibid. and in DeHass and Hollinger, “3D Heritage Preservation.” 74 Hollinger et al., “Tlingit-Smithsonian Collaborations.” 75 Smithsonian, Office of the Chief Information Officer, “Smithsonian Institution Information Technology Plan” (2012) FY 2012–FY2016. 76 The fields that the different departments use are quite different because of the type of content or because of different collections management practices. The registrar’s office and IT office have access to the entire database, including all the fields for all of the departments. 77 Previously, this was owned by KE Software, known as KE EMu. 78 Smithsonian, Office of the Chief Information Officer, “Smithsonian Institution Information Technology Plan” (2012) FY 2012–FY2016, 66.

N o t e s t o pa g e s 1 4 – 2 4

197

79 Smithsonian Institution Annual Reports (hereafter SIAR), 1898, 32. 80 Dauber, “Bureaucratizing the Ethnographer’s Magic,” 75. 81 Bruchac, Savage Kin, 177.

Chapter 1: Writing Desiderata 1 For more excellent discussions, see Anderson, Imagined Com­munities; Appadurai, Modernity at Large; Fujimura, “Authorizing Knowledge in Science and Anthropology”; and Verdon “The World Upside Down.” 2 For more on the establishment of objectivity in the sciences, see Daston and Galison, Objectivity. 3 Bennett, Pasts beyond Memory. 4 The establishment of the ocularcentric approach to objectivity is also historical, and learning through objects or through vision is a process that has been privileged in the sciences and elsewhere. See Daston, “Type Specimens and Scientific Memory.” However, it was not the only, or even most important, approach to understanding objects. As I show in later chapters, the documentary evidence applied to objects and Indigenous knowledge through cataloguing proves just as important as, or even more important than, the objects themselves. 5 Dauber, “Bureaucratizing the Ethnographer’s Magic.” 6 Coote, “Notes and Queries and Social Interrelations”; Petch, “Notes and Queries and the Pitt Rivers Museum.” 7 Dauber, “Bureaucratizing the Ethnographer’s Magic,” 76. 8 Smithsonian Institution Annual Reports (hereafter SIAR), 1847, 1. 9 Nichols, “Museum Networks,” 7. 10 Ibid., 10. 11 Hinsley, Savages and Scientists; SIAR, 1859, 45. 12 Hinsley, Savages and Scientists. 13 SIAR, 1875, 67–68. 14 Ibid. 15 Hinsley, Savages and Scientists; Judd, The Bureau of American Ethnology. 16 Urry, “‘Notes and Queries on Anthropology,’” 45. 17 SIAR, 1854, 79. 18 Roberts, “One Hundred Years of Smithsonian Anthropology,” 120. 19 Joseph Henry, “Circular in Reference to American Archaeology,” Circular no. 316, in Smithsonian Miscellaneous Collections (1878), Smithsonian Institution Archives (hereafter SIA), RU00058, Collected Letters on Ethnology, 1876–79, box 1, folder 1. 20 The first mention found of tribal affiliation in the annual reports is for 1854. However, the collection of other desiderata are also referenced in Spencer Baird, “General Directions for Collecting and Preserving Objects of Natural History” (1848) SIA, Smithsonian Institution Chief Clerk, Forms, Circulars, Announcements, 1846–1933, RU000065, box 1, and in “Indian Languages of North America,” Circular no. 1, June 1852. 21 Parezo, “The Formation of Ethnographic Collections,” 3. 22 Hinsley, Savages and Scientists. Baird was an unpaid collaborator (or correspondent with the Smithsonian) and was working at Dickinson College as a full professor of natural history. He was hired in 1850 as a curator at the National Museum of the Smithsonian and assistant to the secretary. In 1872, he was in charge of the USNM and then was elected secretary of the Smithsonian (1878–87). 23 Nichols, “Museum Networks”; Rivinus and Youssef, Spencer Baird of the Smithsonian. 24 This was reprinted in a later publication called “Directions for Collecting, Preserving, and Transporting Specimens of Natural History” (1852).

198

N o t e s t o pa g e s 2 4 – 3 5

25 Baird, “General Directions for Collecting and Preserving Objects.” 26 Ibid.; emphasis added. 27 Ibid. 28 SIAR, 1851, 25. 29 This is likely Thomas Sidney Jesup (1788–1860), a US Army Officer, who commanded troops during the Second Seminole War (1835–42), and is to be distinguished from Morris Jesup, the philanthropist and president of the American Museum of Natural History, who sponsored the Jesup North Pacific Expedition 1897–1902. 30 Thomas Jesup to Colonel S. Churchill, 31 March 1848, SIA, Smithsonian Institution, Chief Clerk, Forms, Circulars, Announcements, 1846–1933, RU000065, box 1. 31 J. Hampdon Porter, A Bibliography of Anthro-Biology [1904?], Anthropology Manuscripts, National Anthropological Archives (hereafter NAA), series 17, Division of Ethnology Manu­ scripts and Pamphlets file, box 2, folder 21. 32 SIAR, 1851, 25. 33 Nichols, “Museum Networks,” 105. 34 Hochman, Savage Preservation. 35 “Indian Languages of North America,” Circular no. 1, June 1852, SIA, Smithsonian Institution, Chief Clerk, Forms, Circulars, Announcements, RU00065, 1846–1933, box 1. 36 Smithsonian collaborators, curators, and ethnologists were also drafting a grammar and dictionary of the Dakota language at this time, which, as noted in the circular, promised to be a model applicable to “most of the languages of this continent.” Inferring thoughts and wishes from archival letters is always difficult, but it is no exaggeration to say that these guides – in naming the world – had a profound effect on the construction of a mythical past that still infects much popular discourse about non-settler history. 37 SIAR, 1861, 392. The first mention of a field guide or circular in the annual reports is actually that of Baird in the annual report of 1850 on page 25; I could not locate a copy of this original circular. 38 George Gibbs, “Instructions for Research Relative to the Ethnology and Philology of Amer­ ica,” Smithsonian Miscellaneous Collections (1863), 180. Eventually, John Wesley Powell pub­ lished a revision to Gibbs’s guide in 1877, entitled “Introduction to the Study of Indian Languages,” which served as a further expansion of Gibbs’s 1863 guide. 39 Gibbs, “Instructions for Research,” 1. 40 Ibid. 41 Ibid., 2. 42 Reardon and TallBear, “‘Your DNA Is Our History.’” 43 Gibbs, “Instructions for Research,” 3. 44 Ibid., 4. 45 Ibid., 5. 46 Ibid., 4. 47 Ibid., 5. 48 This misunderstanding highlights exactly why Indigenous data sovereignty today is important. For more on Indigenous data politics, see Kukutai and Taylor, eds., Indigenous Data Sovereignty. 49 Gibbs, “Instructions for Research,” 5. 50 Ibid., 7. I am not suggesting that Gibbs’s work has not been important for Indigenous move­ ments. In his career, Gibbs argued for Indigenous fishing and hunting rights in perpetuity, and he understood that burial mounds were connected to living people as well. 51 However, it is difficult to tell if these suggestions were followed by collectors, as the documentation is often very sparse. 52 Gibbs, “Instructions for Research,” 2.

N o t e s t o pa g e s 3 5 – 4 1

199

53 Ibid., 8. 54 Ibid., 6. 55 Ibid. 56 Ibid., 12. 57 Ibid., 13. 58 Joseph Henry to Captain Howard, 27 May 1867, Smithsonian Miscellaneous Collections, SIA, RU 00065, box 2. 59 “Suggestions Relative to Objects of Scientific Investigation for the Expedition under Capt. Howard, along the Coast of Russian America,” Smithsonian Miscellaneous Collections (1867), SIA, RU00065, box 1, 9. 60 The 1867 document is likely a second or third printing, and probably was revised. I found original correspondence dating from 1866 from Joseph Henry to the surgeon general of the US Army that requests that the circular be distributed to medical officers. The circular was republished in 1869. Letter from Joseph Henry to the Surgeon General, NAA, Smith­ sonian Institution, Chief Clerk, Forms, Circulars, Announcements, RU00065, 1846–1933, box 1. 61 Joseph Henry, “Circular Relating to Collections in Archaeology and Ethnology,” Circular no. 205, in Smithsonian Miscellaneous Collections, vol. 8 (1867). 62 Ibid. 63 Ibid., 2. 64 Ibid. 65 Henry Davis, “To the Public: Ethnology, Archology [sic], and Philology of the Races of Mankind Inhabiting Either None or at Any Previous Period the Continent of America” (1867), Smithsonian Miscellaneous Collections, SIA, RU00065, box 1. 66 I was unable to determine through archival research exactly which circulars were present on which expeditions, and it is unlikely this information can be uncovered. 67 SIAR, 1862, 87. 68 Ibid., 88. 69 SIAR, 1876, 35. 70 Ibid., 35. 71 An example is the British “Notes and Queries,” which served a similar purpose for field studies. For more on this, see Urry, “‘Notes and Queries’ on Anthropology.” 72 Nichols, “Museum Networks.” 73 For example, the establishment of the UK Museums Association Conference in 1902. 74 SIAR, 1875, 6. 75 Hinsley, Savages and Scientists, 151. 76 Material collected by the BAE was usually transferred directly to the USNM, until the two departments merged in 1965. 77 Mason, Ethnological Directions. 78 SIAR, 1875, 41. 79 Ibid., 67–68. 80 SIAR, 1878, 30. 81 SIAR, 1862, 88. 82 SIAR, 1875, 67–68. 83 For more on world’s fairs and expositions at this time, see Evans and Boswell, Representing the Nation, and Rydell, All the World’s a Fair. 84 Mason, Ethnological Directions, 3. 85 Ibid., 1. 86 Ibid., 11–13.

200

N o t e s t o pa g e s 4 1 – 4 9

87 Otis T. Mason, Annual Report, 1883–84, SIA, Department of Anthropology, Curator’s Annual Reports, RU000158, series 1, box 3, folder 4 (1884), 3–4. 88 Nichols, “Museum Networks,” 120. 89 Mason, Annual Report, 1883–84, 3. 90 Mason, Ethnological Directions, 3. 91 Ibid., 3. 92 Otis Mason also transcribed a “Code of Symbols for Charts of Prehistoric Archaeology” for the Smithsonian’s Annual Report in 1875. The document was intended to act as a tool for standardizing the reporting on archaeological finds in the United States; earlier code systems had “failed to become popular, because they were understood only in the idiom wherein they were written, or they have been so mixed up with geology on the one hand and with history proper on the other, as to become far too complex for popular use” (SIAR, 1875, 221). 93 Powell, Introduction to the Study of Indian Languages, 3. 94 Ibid. 95 Henry, “Circular in Reference to American Archaeology,” Circular no. 316, Smithsonian Miscellaneous Collections (1878). For Mason’s involvement, see SIAR, 1878, 15. 96 In 1883, Charles Rau, a Smithsonian archaeologist and antiquities collector, drafted a similar circular, which was designed for collectors in the United States. It described the different kinds of archaeological finds that were desired by the Smithsonian. Similarly, in 1884, Cyrus Thomas described the kinds of necessary observances for mound excavation in a circular entitled “Directions for Mound Exploration.” Rau, “Circular Relative to the Contribution of Aboriginal Antiquities” and Thomas, “Directions for Mound Exploration.” 97 Henry, “Circular in Reference to American Archaeology,” 2. It would be interesting to imagine that a map of early collections may reveal a map of early post offices and train stations, instead of locations where these objects were collected. 98 Ibid., 6. 99 Ibid., 4–5. 100 Ibid., 6. 101 Otis T. Mason, “Summary of Correspondence of the Smithsonian Institution Previous to January 1, 1880, in Answer to Circular No. 316,” SIAR, 1879, 428–48. 102 SIAR, 1878, 30–31. 103 Rau, “Circular Relative to the Contribution of Aboriginal Antiquities”; Thomas, “Directions for Mound Exploration”; “Circular,” Canadian Institute” (1889), NAA, Division of Ethnology, Manuscripts and Pamphlets File, box 2, folder 152. 104 Rau, “Circular Relative to the Contribution of Aboriginal Antiquities,” 482. 105 At this time, Holmes was head curator in the USNM’s Department of Anthropology; Otis Mason was curator in the Division of Ethnology, which was a subdivision of the Anthropol­ ogy Department. 106 Holmes and Mason, “Instructions to Collectors,” 3. 107 Ibid., 1. 108 Ibid., 4. 109 Ibid., 5. 110 Ibid. 111 Ibid., 6. 112 Ibid. 113 Parezo, “The Formation of Ethnographic Collections,” 4. 114 Gibbs, “Instructions for Research,” 1. 115 Ibid., 2. 116 Ibid., 3.

N o t e s t o pa g e s 4 9 – 5 7

201

117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134

Daston and Galison, Objectivity. Mason, Ethnological Directions, 4. Gibbs, “Instructions for Research,” 14. Ibid., 11. Ibid., 6; Mason, Ethnological Directions, 4. Mason, Ethnological Directions, 3. Ibid., 4. Bruchac, Savage Kin. Mason, Ethnological Directions, 35. Parezo, “The Formation of Ethnographic Collections”; Parezo, “Cushing as Part of the Team.” SIAR, 1901, 365. Michel Callon, “Domestication of the Scallops.” Bruno Latour, Science in Action. See the following works by Latour: Reassembling the Social; Pandora’s Hope; and “Visualization and Cognition.” Latour, Laboratory Life, 14. Of course, this understanding can be traced back intellectually to others, such as the early work of the philosopher of science François Dagognet. See Dagognet, “Tableaux et langages de la chimie.” Latour, “Where Are the Missing Masses?” Holmes and Mason, “Instructions to Collectors,” 16. Gibbs, “Instructions for Research,” 2.

Chapter 2: On the Margins 1 As Candace Greene notes, this practice has origins that are much earlier in other American museums. Greene, “Material Connections.” 2 Gitelman, Paper Knowledge, 21–22. 3 Greene, “Material Connections,” 152. 4 One notable exception is a master’s thesis written on the subject of museum day-books that specifically examines their form as documentation devices. See Walklate, “The Quotidian and the Bizarre.” 5 Thomas, “The Documentation of the British Museum’s Natural History Collections.” 6 Translation provided by Erik Kwakkel. 7 Greene, “Documentation, Attribution and the Ideal Type.” 8 Baird was given a personal recommendation for the work of assisting in the “arrangement and description of specimens” by John Audubon. Letter from John Audubon, 30 July 1842, SIA, RU7002, series 3, box 14, folder 8. 9 Jan Letowski, “Written Report of the History of Cataloging at the USNM” [2008?], unpublished internal Smithsonian staff document, Department of Anthropology, National Museum of Natural History (hereafter NMNH), Museum Support Center (hereafter MSC). 10 Baird, Catalogue of North American Birds. 11 Spencer F. Baird, 1846, cited in Walsh, “Collections as Currency,” 202. 12 Walsh, “Collections as Currency,” 203. 13 Simon Knell details the growth of collections through time and the changing relationship between amateurs and professionals in the geological sciences, especially as society and interpretive expectations change. See Knell, “Museums, Reality and the Material World.” 14 SIAR, 1889, 8. 15 SIAR, 1857, 50. 16 Commodore Perry commanded the US Japan Expedition to initiate trade between the two countries, which culminated with the Kanagawa Treaty in 1854.

202

N o t e s t o pa g e s 5 7 – 7 2

17 This is also similar to Mason’s contention that objects were useful as type specimens only when put into context with other objects from around the world, as noted in Otis Mason’s “The Occurrence of Similar Inventions in Areas Widely Apart.” 18 Bowker and Star, Sorting Things Out. 19 Stoler, Along the Archival Grain. 20 Hinsley, Savages and Scientists. 21 This letter is reprinted in a later archival document: Charles Ellis to Robert Elder, Professor Edward Foreman, and the Amer. Bureau of Ethnology, Illustrator of Dept. of Anth. Catalogs (1973), unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 22 Daston, “Type Specimens and Scientific Memory.” 23 Moser, “Making Expert Knowledge through the Image,” 59. 24 For an example of how field-note drawings were used in ethnographic work and are important for communities today, see Glass, “Drawing on Museums.” 25 This is not unlike what happened when the first photographic images of objects were produced in the museum – they seem to have existed only for quick visual reference and were not viewed as a systematic object-documentation approach until more recently. 26 Nikolova-Houston, “Margins and Marginality.” 27 Landow, HyperText. 28 Haas, “Wampum as Hypertext.” 29 Hollinger, Assessment of Tlingit Objects. 30 Timothy Dixon Bolles to Smithsonian Institution, 16 July 1884, NMNH, Accession Record 15153, Registrar’s Office, Department of Anthropology, Smithsonian Institution. 31 Timothy Dixon Bolles to Smithsonian Institution, 3 November 1884, ibid. 32 Otis T. Mason, Annual Report, 1884–85, SIA, Department of Anthropology, Curator’s Annual Reports, RU000158, series 1, box 3, folder 5, 3–4. 33 Mason, Annual Report, 1885–86, SIA, Department of Anthropology, Curator’s Annual Reports, RU000158, series 1, box 3, folder 6, 1886. 34 Timothy Dixon Bolles, “A Preliminary Catalogue of the Eskimo Collection in the U.S. National Museum, Arranged Geographically and by Uses,” SIAR, 1887, 335–65. 35 Ibid., 335. 36 Mason, Annual Report, 1885–86. 37 Bolles, “A Preliminary Catalogue,” 335. 38 It seems that Bolles may have included his own collection in his catalogue as well. He lists a small number of “Thlinkiit” objects; however, these may be only a fraction of the total sixty-eight objects collected from Hoonah. 39 Hoonah Repatriation Collection, Accession 15153, Smithsonian Department of Anthropology, Museum Support Center, Suitland, MD; Hollinger, Assessment of Tlingit Objects. 40 Ensign Albert P. Niblack, “The Coast Indians of Southern Alaska and Northern British Columbia,” in SIAR, 1888, 272. 41 Ibid. Niblack notes that this classification scale is not sufficient and states: “It might be well to describe here all the accessories of ceremonial costumes, such as the accompanying rattles, snappers, drums, whistles, etc. These, however, are reserved for Chapter VII, where they are dealt with as musical instruments” (272). 42 Ibid., 332. 43 According to Hollinger, Assessment of Tlingit Objects, there is a Tlingit dictionary that lists this term for rattles shaped like skulls, but it may not be the term used by the Tlingit people themselves to describe these rattles. There is nothing in Bolles’s documentation that suggests he used the term originally; it was applied later by a museum worker in the process of documentation.

N o t e s t o pa g e s 7 4 – 8 3

203

44 Hollinger, Assessment of Tlingit Objects, 23. 45 Nichols, “Lost in Museums.” 46 As he writes, “Fearing to send what you might consider old rubbish and of little value yet I feel that they are of value if you have not already too many.” Timothy Dixon Bolles to Smith­ sonian Institution, 16 July 1884, NMNH, Accession Record 15153, Registrars Office, MSC. 47 These were likely exchanged with the Harvard Peabody Museum and the Leiden Museum. 48 J. Hampdon Porter, A Bibliography of Anthro-Biology, 33. 49 A typical claim of early salvage ethnology. 50 Daston and Galison, Objectivity.

Chapter 3: Ordering Devices and Indian Files

1 Fenton, “The Museum and Anthropological Research.” 2 Foucault, Discipline and Punish, 281. 3 Krajewski, Paper Machines, 10. 4 Wright, Cataloging the World. 5 Krajewski, Paper Machines, 13. 6 Ibid., 70. 7 Ibid., 82. 8 However, as Krajewski notes, Jewett idealized a stereotyped clay-printing plates system that never saw the light of day. Ibid., 104. 9 Ibid., 90. 10 Meyer, Studies of the Museums and Kindred Institutions. 11 Ibid. 12 Urban, “Library Influence on Museum Information Work.” 13 Krajewski, Paper Machines, 4. 14 SIAR, 1881, 96. 15 SIAR, 1880, 56. 16 Mason published nearly eighty articles and studies based on Smithsonian collections, including his widely cited works Aboriginal American Basketry and The Origins of Invention. 17 For more on Boas, see Regna Darnell, And Along Came Boas: Continuity and Revolution in Americanist Anthropology (Philadelphia: John Benjamins Publishing, 1998). 18 Mason, Annual Report, 1905–6, SIA, Department of Anthropology, Curator’s Annual Reports, RU000158, series 2, box 26, folder 17. 19 Bennett, Pasts beyond Memory; Foucault, The Order of Things; Hooper-Greenhill, Museums and the Shaping of Knowledge. 20 For more, see Hinsley, The Smithsonian and the American Indian; Buettner-Janusch, “Boas and Mason”; Kohlstedt, “Otis T. Mason’s Tour of Europe”; and Nichols and Parezo, “Social and Material Connections.” 21 Mason, Annual Report, 1884–85, SIA, Department of Anthropology, Curator’s Annual Reports, RU000158, series 1, box 3, folder 5. The person responsible for this severe reckoning remains unnamed in the documentation, but Miss Sarah Latham’s work is felt throughout the ledgers and catalogue cards. 22 The Diary of Otis T. Mason, entry for 1 July 1884, National Anthropological Archives (hereafter NAA), file: Anthropology, History of, 4903. 23 Otis T. Mason, Annual Report, 1882–83, SIA, Department of Anthropology, Curator’s Annual Reports, RU000158, series 1, box 3, folder 3. 24 There is a terminological distinction between catalogue cards and the descriptive museum catalogues: the latter were books that documented an entire collection or exhibition. For example, collectors in the field created catalogues of the collection they brought into the

204

N o t e s t o pa g e s 8 4 – 9 4

museum. The ledgers are also referred to as the museum “register” or the “Ethnological Cata­log.” They were not intended to provide major contextual information about the objects but simply to list them as accessions. 25 Otis T. Mason, Annual Report, 1905–6, SIA, Department of Anthropology, Curator’s Annual Reports, RU000158, series 2, box 26, folder 17. 26 Mason describes this process with a specific example of a Hawaiian basket: “A most rare twined Hawaiian basket was only a few weeks ago bought for a California example by a tourist, among a former missionary’s effects in Florida, and taken to Milwaukee. By the rarest piece of good fortune it was shown in the National Museum at the moment when Bringham’s memoir on baskets of the ancient Hawaiians was being studied ... and a bit of false identification prevented. This incident could be matched a great many times ... To obtain the best results, fraud and error must be vigorously eliminated from the science. The advantage of exploration by educated and trained experts will be mentioned later.” Mason, Annual Report, 1905–6, 3. 27 SIAR, 1884, 12; see also Nichols, “Museum Networks.” 28 SIAR, 1884, 12. 29 Mason, Annual Report, 1885–86, SIA, Department of Anthropology, Curator’s Annual Reports, RU000158, series 1, box 3, folder 6. Mason’s annual curator’s reports document increasingly standard and rigorous methods for the procedures of accessioning and cataloguing. 30 SIAR, 1884, 12. 31 In the Smithsonian annual report for 1884, a separate section entitled the “Principles of Administration” notes that, in order to make objects accessible to the public, in a clearly classificatory arrangement, “the objects should be so carefully classified that their relations to each other may be recognized by the visitor so that taken together, they shall suggest general conclusions; in the formation of these conclusions he should be aided by certain general or collective labels which relate to and describe groups of objects in a manner similar to that in which the individual labels describe separate articles.” See SIAR, 1884, 12. 32 Ibid. 33 Mason, Annual Report, 1885–86. 34 Ibid. 35 Ibid. 36 Mason, Annual Report, 1884–85. 37 Ibid. 38 Mason, Annual Report, 1885–86. 39 Genoways and Andrei, Museum Origins. 40 SIAR, 1893, 36. 41 Mason, Annual Report, 1885–86. 42 Ibid. 43 For his account of this, see Mason, Annual Report, 1894–95, SIA, Department of Anthropol­ ogy, Curator’s Annual Reports, RU000158, series 1, box 3, folder 15. 44 Hinsley has argued that, according to Klemm, “the ethnologist must follow the lead of the naturalist and analyze his subject in all its geographical and developmental variety, then put all his observations together to present the larger historical picture.” Hinsley, Savages and Scientists, 88. 45 SIAR, 1890, 119. 46 Kohlstedt, “Otis T. Mason’s Tour of Europe”; Nichols and Parezo, “Social and Material Connections.” 47 For example, the monographic catalogues were based on the plan published by the South Kensington Museum: “monographic catalogues, based on the plan of those published by the South Kensington Museum, have also been commenced” (SIAR, 1891, 25). It should be noted that these are not the same as the encyclopedic card catalogues that Mason is referring

N o t e s t o pa g e s 9 5 – 9 9

205

to; the monographic catalogues are more akin to exhibition catalogues inspired by art history approaches, rather than the systematic card catalogues that were in libraries. 48 Mason, Annual Report, 1884–85. 49 Otis T. Mason, 1 July 1884, Diary, file: Anthropology, History of, NAA, 4903. 50 Ibid., 5 July 1884. 51 Mason, Ethnological Directions, 24. 52 Hooper-Greenhill, Museums and the Shaping of Knowledge, 13. 53 Ibid. 54 Foucault, The Order of Things. 55 As Foucault observes, the classical episteme was fundamentally rejected by the end of the seventeenth century, ibid., 218. 56 Nichols, “Museum Networks,” 288. 57 Mason, Annual Report, 1891–92, SIA, Department of Anthropology, Curator’s Annual Reports, RU000158, series 1, box 3, folder 12. 58 Otis T. Mason, Annual Report, 1896–97, SIA, Department of Anthropology, Curator’s Annual Reports, RU000158, series 2, box 4, folder 2. 59 Mason, Annual Report, 1899–1900, SIA, Department of Anthropology, Curator’s Annual Reports, RU000158, series 2, box 26, folder 12. 60 SIAR, 1902, 57. 61 J.W. True to William Henry Holmes, 12 January 1900, NAA, series 2, Administrative Series File, Department of Anthropology Subject File, Anthro. Misc. 62 F.A. Bather, “Chicago Museum,” 296. 63 Reference to the individual catalogues of exchanges, articles under conservation treatment, and duplicate materials are sparse after this point. Other recording books may have fallen into disuse after Mason’s death in 1908 as the card catalogue, used in conjunction with the large ledger books of accessions, was adopted widely throughout the museum. 64 Otis Mason, Annual Report, 1904–5, SIA, Department of Anthropology, Curator’s Annual Reports, RU000158, series 2, box 26, folder 16, 56. 65 Mason, Annual Report, 1906–7, SIA, Department of Anthropology, Curator’s Annual Re­ ports, RU000158, series 2, box 26, folder 17. 66 Ibid. 67 Ibid. 68 SIAR, 1905, 30. 69 Mason, Annual Report, 1907–8, SIA, Department of Anthropology, Curator’s Annual Re­ ports, RU000158, series 2, box 26, folder 19. 70 As Mason notes in his 1908 annual report: “In addition to the consecutive listing and numbering of specimens, the card cataloguing must be carefully done and data secured for labels and monographs. To avoid confusion and loss, the collections must find their proper places in exhibition, study, and reserve series. Anthropological materials demand extraordinary care in storage. In addition to the traditional enemies of moth, rust, and thieves, a great deal of it cannot be bent or folded and takes up a vast amount of space.” Ibid. 71 SIAR, 1908, 40. 72 SIAR, 1912, 30. 73 SIAR, 1945, 28. 74 Office Memoranda for FM Setzler, 3 June 1952, Work Performance of Mrs. Hilda J. Davis, NAA, Administrative Series File, Department of Anthropology Subject File, box 73, Eth­ nology, 1930–1961. 75 This is articulated clearly in Richard Urban’s “Library Influence on Museum Information Work.”

206

N o t e s t o pa g e s 9 9 – 1 0 6

76 For example, see England and Boyer, “Women’s Work”; Kittler, Gramophone, Film, Typewriter; Muhr and Rehn, “On Gendered Technologies and Cyborg Writing.” 77 Miss Sarah E. Latham was a copyist at the Smithsonian from 1883 to 1887, and then a clerk from 1889 in the official register of the museum, and it is shown that she began work in the United States Civil Service Registers, 1883–1887. She donated a small coin to the museum in 1887, SIAR, 1921–22. I am indebted to Kelsey Falquero for providing the research for this much-needed biographical note. 78 Saul Riesenberg to Frank M. Setzler, Summer Interns, 24 February 1959, NAA, Administrative Series File, Department of Anthropology Subject File, box 73, Ethnology, 1930–1962, 1959. 79 Saul Riesenberg to A.C. Smith, Request for Application of Temporary Museum Aide, 3 October 1959, NAA, Administrative Series File, Department of Anthropology Sub­ject File, box 73, Ethnology, 1930–1961. 80 SIAR, 1908, 40. 81 Walter Hough, Report of the Department of Anthropology, 1920–21, NAA, series 14, Annual Reports 1920–1983, box 119, folder 1920–1921, 7. 82 Ibid., 16. 83 Hough, Report of the Department of Anthropology, 1923–24, NAA, series 14, Annual Reports 1920–1983, box 119, folder 1923–24. 84 SIAR, 1923, 14. 85 Hough, Report of the Department of Anthropology, 1922–23, NAA, Series 14, Annual Reports 1920–1983, box 119, folder 1922–23, 15. 86 Hough, Report of the Department of Anthropology, 1924–25, NAA, series 14, Annual Reports 1920–1983, box 119, folder 1924–25. 87 Hough, Report of the Department of Anthropology, 1925–26, NAA, series 14, Annual Reports 1920–1983, box 119, folder 1925–26. 88 Hough, Report of the Department of Anthropology, 1928–29, NAA, series 14, Annual Reports 1920–1983, box 119, folder 1928–29. 89 H.W. Krieger to Mr. Collings, Memorandum for Mr. Collings, 15 October 1936, NAA, Ad­ministrative Series File, Department of Anthropology Subject File, box 73, Ethnology, 1930–1961. 90 H.W. Krieger to Frank Setzler, Memorandum for Mr. Setzler, 24 December 1935, NAA, Accessioning and Cataloguing of Specimens in the Division of Ethnology, Admin­istrative Series File, Department of Anthropology Subject File, box 73, Ethnology, 1930–1961. 91 H.W. Krieger to Frank Setzler, Memorandum for Mr. Setzler, n.d., NAA, Adminis­tra­tive Series File, Department of Anthropology Subject File, box 73, Ethnology, 1930–1961. 92 Ibid. 93 H.W. Krieger to Frank Setzler, Memorandum for Mr. Setzler, 20 February 1939, NAA, Division of Ethnology, Administrative Series File, Department of Anthropology Sub­ject File, box 73, Ethnology, 1930–1961. 94 H.W. Krieger to Frank Setzler, Memorandum for Mr. Setzler, 24 December 1935, NAA, Admin­istrative Series File, Department of Anthropology Sub­ject File, box 73, Ethnology, 1930–1961. 95 Ibid. 96 Charles Wetmore to H.W. Krieger, 7 January 1936, NAA, series 2, Alpha Subject File, 1828– 1963, Accession Routine. 97 Anon, Note [1936?], NAA, series 2, Alpha Subject File, 1828–1963, Accession Routine. 98 Walter Hough, Report of the Department of Anthropology, NAA, series 14, Annual Reports 1920–1983, box 120, folder 1936–37. 99 Anon, Note.

N o t e s t o pa g e s 1 0 6 – 1 3

207

100 Frank Setzler, Department of Anthropology Annual Report, 1939–40, NAA, series 14, Annual Reports 1920–1983, box 120, folder 1939–40. 101 Ibid. 102 Ibid. 103 Frank Setzler, Department of Anthropology Annual Report, 1942, NAA, series 14, Annual Reports 1920–1983, box 120, folder 1941–42. 104 Ibid. 105 Setzler, Department of Anthropology Annual Report, 1943, NAA, series 14, Annual Reports 1920–1983, box 121, folder 1942–43. 106 Setzler, Department of Anthropology Annual Report, 1946, NAA, series 14, Annual Reports 1920–1983, box 121, folder 1946–47. 107 Accession Routine, n.d., NAA, series 2, Alpha subject, 1828–1963, box A, folder “Accession Routine.” 108 Frank Setzler, Department of Anthropology Annual Report, 1951, NAA, series 14, Annual Reports 1920–1983, box 121, folder 1950–51. 109 Ibid. 110 SIAR, 1880, 58. 111 Riesenberg to Setzler, Request for [Temp] Work Typist. 112 Ibid. 113 Frank Setzler, Department of Anthropology Annual Report, 1954, NAA, series 14, 1920–1983, box 122, folder 1954–55. 114 SIAR, 1889, 8. 115 SIAR, 1895, 91. 116 SIAR, 1904, 66. 117 Note, 1976, Musical Instruments Card Index, Anthropology Card Index Room, NMNH, MSC. 118 Ibid. 119 Krajewski, Paper Machines.

Chapter 4: Pragmatic Classification 1 Law, “Actor Network Theory and Material Semiotics,” 146; Law, “On Power and Its Tactics.” 2 For more, see Merrill, “William Curtis Sturtevant, Anthropologist,” and Sturtevant, “Studies in Ethnoscience.” 3 Sturtevant, “Studies in Ethnoscience,” 100. 4 Ibid., 103. 5 Ibid., 112. 6 Sturtevant, Guide to Field Collecting of Ethnographic Specimens, 1. 7 Ibid. 8 Ibid. 9 Ibid., 34. 10 Ibid., 11. 11 Ibid., 29. 12 Sturtevant cites the original publication: Tom F.S. McFeat, “The Object of Research in Museums.” 13 Sturtevant, Guide to Field Collecting of Ethnographic Specimens, 29. 14 Ibid. 15 Ibid., 11. 16 Ibid., 7.

208

N o t e s t o pa g e s 1 1 4 – 3 1

17 Ibid. 18 I will not go into detail on the history of the series here; but it is important to note that a newly revised version of the introductory volume of the HNAI will be published in 2020, edited by Igor Krupnik. Many chapters in the new version focus on the change in the discipline and museum work since the 1970s. 19 More recently, these have included individual community names, where possible, and Native language ethnonyms. Candace Greene, “Revisions in Tribal Nomenclature,” 28 February 1996, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 20 The three parts of the National Parks Service’s Museum Handbook are available online, at http://www.nps.gov/museum/publications/handbook.html. The Human Relation Area Files (HRAF) is a large documentation project, now a non-profit organization and research agency based at Yale University. It originated with the work of George Murdock, with his “Outline of World Cultures” and cross-cultural survey projects. It was Murdock’s goal to create a large index of similarities and differences between cultural groups globally. See https://hraf.yale.edu/. 21 Candace Greene, interview with the author, 8 November 2013, Suitland, MD. 22 In 1965, the University of Oklahoma received funding to plan a conference and a pilot study in surveying the ethnological collections that existed in Oklahoma museums, with the goal of entering them into a computer system and creating data standards. For more on the history of indexing and computerization attempts broadly, see Ricciardelli, A Pilot Study, and Turner and Greene, “Access to Native Collections.” 23 William Sturtevant, “Suggested Coding System,” 28 April 1965, NAA, Department of An­ thropology, Admin Records, 1972–94, box 97, folder: Sturtevant’s Museum Catalog Clas­ sification Project. 24 Ibid. 25 “Recommendations for Catalog Descriptions,” n.d., NAA, Department of Anthropology, Admin Records, 1972–94, box 87, file: Ethnology Catalog, Computerization of. 26 Frank Taylor, Accessioning Procedure, 28 February 1967, NAA, Department of Anthropology, Admin Records, 1972–94, box 86, file: Cataloging forms used elsewhere. 27 William Fitzhugh to John Ewers, n.d., re: Computerizing the Ethnological Collections, NAA, Department of Anthropology, Admin Records, 1972–94, box 87, file: Ethnology Catalog, Computerization of. 28 Gordon Gibson to Saul Riesenberg, Description of Ethnological Specimens for the Catalog, 26 January 1968, NAA, Department of Anthropology, Admin Records. 1972–94, box 87, file: Ethnology Catalog, Computerization of. 29 Ricciardelli, A Pilot Study for Inventorying Ethnological Collections. Ross Parry also gives an over­view of some of these processes in his book Recoding the Museum, 22. 30 Gibson to Riesenberg, Description of Ethnological Specimens. 31 Ricciardelli, A Pilot Study for Inventorying Ethnological Collections, 18. 32 Clifford Evans, Requests to be carried out by curatorial museum specialists, technicians and aides, 16 February 1965, Elder Memo Work Requests, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC, 1. 33 Ibid., 3. 34 Parry, Recoding the Museum, 22. 35 Cutbill, Hallan, and Lewis, “A Format for the Machine Exchange of Museum Data”; Roberts and Light, “Progress in Documentation”; Vance, “Museum Computer Network.” 36 Parry, Recoding the Museum, 51. 37 Ibid., 15. 38 Ibid., 16.

N o t e s t o pa g e s 1 3 2 – 3 8

209

39 Chenhall, Museum Cataloging in the Computer Age, 94. 40 Parry, Recoding the Museum, 16. 41 Ethnological collections were considered to be very different from, although similar and connected to, art collections and historical collections. 42 Perhaps the most definitive volume on the subject is Robert Chenhall’s Museum Cataloging in a Computer Age, which outlines the ways in which documentation was talked about and understood during the advent of computerization. Published in 1975, the book uses case studies to document various ways that museums were retrieving and storing collections informa­ tion digitally. Chenhall’s guide, which was used by many museums, outlined the importance of considering the information requirements of the collection prior to computerization. He argued that the kind of computer system and the software programs chosen would depend on the museum and the type of audiences that wished to access the information. 43 Vince Wilcox, Annual Report, 1977–78, NAA, Department of Anthropology, series 14 Annual Reports, 1920–1983, box 126, folder 78. 44 Douglas Ubelaker, Memorandum for Staff, Procedures for Inventory, 16 November 1978, Elder Memorandums, 1978, unpublished internal Smithsonian staff document, Depart­ ment of Anthropology, NMNH, MSC, 2. 45 Robert Elder, “Report on the Cheyenne-Arapaho Inventory,” Project, 1 February–28 April 1978, Elder Cheyenne-Arapaho Inventory File, unpublished internal Smithsonian staff docu­ment, Department of Anthropology, NMNH. 46 Vince Wilcox, “Proposal: Collections Management Program for Inventory Control,” unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC, 1978, 2. 47 Ibid. 48 Ibid., 4. 49 Vince Wilcox, Annual Report, 1979, NAA, Department of Anthropology, series 14 Annual Reports, 1920–1983, box 126, folder 79. 50 This was a standard method of input for IBM typewriters at the time. 51 The Entrex was a type-to-tape computer used for data entry and validation. 52 Wilcox, Annual Report, 1977–78, 3. 53 Wilcox, “Proposal: Collections Management Program for Inventory Control.” 54 Wilcox, Annual Report, 1977–78, 3. 55 Candace Greene, interview with the author. 56 Smithsonian staff member, interview with the author, 8 November 2013, Suitland, MD. 57 Wilcox, Curator’s Annual Report, 1977–78, 3. 58 Candace Greene, interview with the author. 59 Wilcox, “Proposal: Collections Management Program,” 3. 60 Chenhall, Museum Cataloging in the Computer Age, 12 and 14. 61 Ibid., 18. 62 William Sturtevant, Notes HRAF, 1984, Sturtevant Papers, NAA, series 1, box 44, HRAF. 63 Doyle, “Naming, Claiming, and (Re)Creating,” 27. 64 “Introduction to the Use of Index Terms,” n.d., “Guide to Index Terms,” Candace Greene, Smithsonian Staff Documents, Department of Anthropology, NMNH. 65 Mason, Annual Report, 1884–85. 66 Gordon Gibson to William Fitzhugh, Subject: Computerization of the Ethnological Collections, 2 May 1975, NAA, Department of Anthropology, Admin Records, 1972–94, box 87, file: Ethnology Catalog, Computerization of, 1975. 67 William Fitzhugh to Curatorial Staff, HRAF Notation, August 1979, NAA, Department of Anthro­pology, Admin Records, 1972–94, box 87, file: Ethnology Catalog, Computerization of, 1979.

210

N o t e s t o pa g e s 1 4 0 – 5 0

68 Candace Greene and William Merrill to Adrienne Kaeppler, “Funds to Standardize Object Names in the North American Ethnology Collection,” 14 April 1987, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 69 Many authors detail the history of classification and belonging, particularly with respect to ancestor remains. See, especially, TallBear, Native American DNA; Reardon and Tallbear, “‘Your DNA Is Our History’”; and Watkins, “Becoming American or Becoming Indian?” 70 Parry, Recoding the Museum, 25. 71 Candace Greene, “Revisions in Tribal Nomenclature,” 28 February 1996, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 72 Thomas Kavanagh to Candace Greene, re Formalization of procedures for editing/standardizing catalog index terms, n.d., Letters regarding Data Standardization, unpublished internal Smith­ sonian staff document, Department of Anthropology, NMNH, MSC. 73 Thomas Kavanagh to Candace Greene, Memorandum, 2 February 1987, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 74 Thomas Kavanagh to Candace Greene, Memorandum, 2 February 1987. 75 Thomas Kavanagh to Candace Greene, Johanna Humphrey, and William Merrill, Index Pro­ject Plan, 7 January 1988, Notes, Data Standards Rules, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 76 Thomas Kavanagh to Candace Greene, William Merrill, and Johanna Humphrey, re Name rules, 3 February 1988, Notes, Data Standards Rules, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 77 Ibid. 78 Candace Greene, Requisition for Supplies, 11 April 1989, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 79 Candace Greene, interview with the author, 11 September 2013, Suitland, MD. 80 Candace Greene to Collections Staff, North American Index Term Standardization, Index Term Standards, 7 June 1996, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC.

Chapter 5: Object, Specimen, Data 1 Bowker and Star, “Invisible Mediators of Action,” 147. 2 Lampland and Star, Standards and Their Stories. 3 National Museum of the American Indian Act, amended 1996, Pub. L. 104-278, 9 October 1996; the Native American Graves Protection and Repatriation Act, Pub. L. 101-601, 25 U.S.C. 3001 et seq., 104 Stat. 3048, 16 November 1990. 4 Lampland and Star, Standards and Their Stories, 25. 5 These include standards like the metadata standard the Dublin Core, as well as the ISO Z39.19 Standard, the Getty Art and Architecture Thesaurus (GAAT), and Library of Congress Subject Headings (LCSH), as well as user-defined thesauri. 6 Bowker and Star, Sorting Things Out, 151. 7 Examples include staff files such as editing policies and more. One example is “EMu Editing Policies,” 24 August 2005, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 8 “Anthropology Data Manual” (2012), unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC, 34. 9 Ibid. 10 Ibid. 11 George Phebus, Curator’s Annual Report, 1971–72, NAA, Department of Anthropology, series 14 Annual Reports, 1920–1983, box 125, 1972.

N o t e s t o pa g e s 1 5 0 – 6 3

211

12 “Anthropology Data Manual,” 5. 13 Candace Green, Funky Data Problems, EMu Meeting on Priorities, EMu meeting notes and comments, 2007, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 14 For example, in the case of the Mukurtu CMS and Wumpurrarini-kari Archive, see Christen, “Archival Challenges and Digital Solutions in Aboriginal Australia.” The database organizes material based on familial connection rather than type of object. Similar attitudes to organizing information for Indigenous-specific needs and social relationships are outlined in Geismar and Mohns, “Social Relationships and Digital Relationships.” See also Christie, “Words, Ontologies, and Aboriginal Databases”; Lyons et al., “Sharing Deep History as Digital Knowledge.” 15 Robert Elder, “Report on the Cheyenne-Arapaho Inventory Project,” 1 February–28 April 1978, Elder Cheyenne-Arapaho Inventory File, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC, 2. 16 Ibid. 17 Candace Greene, interview, 11 September 2013. 18 William Fitzhugh to John Ewers, Re: Computerizing the Ethnological Collections, n.d., NAA, Department of Anthropology, Admin. Records, 1972–94, Ethnology Catalog, Com­ puterization of. 19 Gordon Gibson, Computerization of the Ethnology Catalog: Understandings Reached in Conference, 8 March 1977, NAA, Department of Anthropology Record Group, box 87, Admin. Records, 1972–94, 1977. 20 Elder, Report on the Cheyenne-Arapaho Inventory, 2. 21 Greene, interview, 11 September 2013. 22 Elder, Report on the Cheyenne-Arapaho Inventory. 23 Vince Wilcox, Curator’s Annual Report, 1979–80, NAA, Department of Anthropology, series 14, Annual Reports 1920–1983, box 127, folder 80, 1980. 24 Vince Wilcox, “Projected Computer Programming Support for Collections Management” (n.d.), NAA, Admin Records, 1972–94, box 97, folder: Sturtevant. 25 Ibid. 26 Beauchamp, email to the author. 27 Smithsonian staff member interview, 17 September 2013. 28 Ibid. 29 Ibid. 30 “Culture Terms Not in Use” (2012), unpublished internal Smithsonian staff document, De­ partment of Anthropology, NMNH, MSC. 31 Further, terms such as “Chink” and “Jap” were also used historically, but were so rare they were not included in this document. Smithsonian staff member interview. 32 “Culture Terms Not in Use.” 33 Ibid. 34 The National Museum of the American Indian, which was finally inaugurated in 2004, holds the former Museum of the American Indian Gustav Heye Collection (transferred to the Smithsonian in 1990) and exhibits new collections of artwork and belongings from many Native American communities. It is important to note that the historical ethnographic collections from the beginning of the Smithsonian Institute are all held in the NMNH – the focus of this book. For more on the history of the NMAI, see Atalay, “No Sense of the Struggle”; Cobb, “The National Museum of the American Indian”; Conn, “Heritage vs. History”; Isaac, “What Are Our Expectations Telling Us?”; Jacknis, “A New Thing?”; Lonetree, “Missed Opportunities”; and Rosoff, “Integrating Native Views into Museum Procedures.” 35 Quoted in Hollinger, Assessment of Tlingit Objects, 4.

212

N o t e s t o pa g e s 1 6 3 – 7 3

36 Despite these requirements, the process of discovering what objects are in museum collections can be difficult for Native American communities, owing to the fact that they rarely have access to field notes or “inside” knowledge of collecting practices (Bruchac, Savage Kin). 37 “What Is Repatriation?” Smithsonian, National Museum of Natural History, online at http:// anthropology.si.edu/repatriation/whatis/. 38 Candace Greene, to collections staff, North American Index Term Standardization, Index Term Standards, 7 June 1996, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 39 C. Smythe, Johanna Humphrey, and Candace Greene, “Summary of Meeting on Repatria­ tion Related Data Issues,” 21 and 22 November 1995, Repatriation Data Issues, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 40 At this time, both INQUIRE and Microsoft Access were used as key databases to contain the collection. 41 It should be noted that, today, when conducting research, the Repatriation Office almost never relies on an object name as a primary field. Objects are found in the database either by geographical location or by the accession number associated with a specific donor. 42 Candace Greene, “Standardization of North American Nomenclature,” Greene to Collections Staff, 4 June 1996, unpublished internal Smithsonian staff document, Department of An­ thropology, NMNH, MSC. 43 Candace Greene, Requisition for Supplies, 11 April 1989, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 44 Greene, “Standardization of North American Nomenclature.” 45 Report of the 16 March 2007 Meeting of the Anthropology EMu Advisory Committee, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 46 Saul Riesenberg to All Personnel, Division of Ethnology, 13 August 1962, Accession Memo­ randa, NAA, Admin. Records, 1972–94, box 87, Ethnology Catalog, Computerization of. 47 Greene, “Standardization of North American Nomenclature.” 48 Carrie Beauchamp to Candace Greene, Re Anthropology Database (EMu) on the Web, 4 December 2006, Email Correspondence, EMu Correspondence 2006, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 49 Kim Neutzling to Staff, YODA to EMu, 27 September 2006, Email Correspondence, EMu Correspondence 2006, unpublished internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 50 Report of the 16 March 2007 Meeting of the Anthropology EMu Advisory Committee, un­published internal Smithsonian staff document, Department of Anthropology, NMNH, MSC. 51 Ibid. 52 Star, “The Ethnography of Infrastructure.” 53 Taken from the NMNH Data Access Policy, “Terms and Conditions for Use of Online Collections Databases,” at https://naturalhistory.si.edu/rc/db/2data_access_policy.html. There is also a much broader data policy for the institution for records data and access, known as Smithsonian Directive 609, published in 2011. This document outlines specific categories of information and images that can be restricted; all other assets are supposed to be accessible to all for non-commercial purposes or “fair use.” Physical anthropology falls under the Native American human remains category, as well as, in many cases, the personally identifiable category. 54 Douglas, Purity and Danger, 44. 55 John Law, “Notes on the Theory of the Actor-Network,” 387.

N o t e s t o pa g e s 1 7 3 – 8 2

213

Conclusion: A Museum Data Legacy for the Future 1 Povinelli, “The Woman on the Other Side of the Wall,” 158. 2 Hays-Gilpin and Lomatewama, “Curating Communities at the Museum of Northern Arizona.” 3 These issues have been raised elsewhere. For example, see the following work in ontology, digital archives, and collaborative cataloguing: Bell, Christen, and Turin, “Introduction”; Christie and Verran, “Digital Lives in Postcolonial Aboriginal Australia”; Doyle, “Naming, Claiming, and (Re)Creating”; Hunt, “Ontologies of Indigeneity”; Krmpotich and Peers, This Is Our Life; and Todd, “An Indigenous Feminist’s Take.” 4 There are many examples, but an important one is the recent report on the restitution of African cultural heritage in France: see Sarr and Savoy, The Restitution of African Cultural Heritage. 5 For an example of this approach in an African museological context, see Laura Gibson, “De­ colonising South African Museums in a Digital Age.” 6 Bowker and Star, Sorting Things Out, 153. 7 Krajewski, Paper Machines. 8 Here I am calling toward the theorization of structure and change from Star and Ruhleder’s work on infrastructure and design. For more, see Star and Ruhleder, “Steps Toward an Ecology of Infrastructure,” 111. 9 For more examples of encounters with legacy data in other archival contexts, see James Opp, “The Colonial Legacies of the Digital Archive.” 10 Bowker and Star, “How Things (Actor-Net) Work,” 199. 11 For more on infrastructures, see Bowker and Star, “Invisible Mediators of Action.” 12 The NMNH is currently in the process of digitizing all of the anthropology accession files – an important step. 13 See Hollinger et al., “Tlingit-Smithsonian Collaborations,” and more recent work in DeHass and Hollinger, “3D Heritage Preservation.” 14 Geismar, Museum Object Lessons for the Digital Age, 111. 15 Here I reference the work of artists Nora Al-Badri and Nikolai Nelles and their project “The Other Nefertiti,” see http://nefertitihack.alloversky.com/. A useful discussion of this project appears in Geismar, Museum Object Lessons for the Digital Age. 16 This is something I am exploring with the collaborative project “Wrapped in the Cloud,” which is a digitization of a weaving by the artist Meghann O’Brien (Jaad Kuujus), with Jaimie Isaac, Kate Hennessy, and Conrad Sly. We have produced a digitized 3D model of a Raven’s Tail-Chilkat combination robe to enable its return to Meghann’s community and family. See Turner et al., “New Media.” 17 For more on reproductions and objects-as-information, see Isaac, “Whose Idea Was This?” and Latham, “Museum Object as Document.” 18 See Turner et al., “New Media.” 19 Wemigwans, A Digital Bundle. 20 Many of these ideas were inspired by my time as a research assistant to help develop the Reciprocal Research Network (www.rrncommunity.org), where we worked through and raised many issues in how to map community values onto existing museum catalogue ontologies. For more, see Susan Rowley, “The Reciprocal Research Network.” 21 For examples of contemporary digital collections projects that focus on specific communities of practice, see the Great Lakes Research Alliance for the Study of Aboriginal Arts and Culture (GRASAC), http://grasac.org; the Reciprocal Research Network (RRN), www. rrncommunity.org; the Mukurtu CMS, http://www.mukurtu.org/; and the Digital Sq’éwlets Project, http://digitalsqewlets.ca/index-eng.php. For more, see Christen, Merrill, and Wynne,

214

N o t e s t o pa g e s 1 8 5 – 9 2

“A Community of Relations”; Lyons et al., “Sharing Deep History as Digital Knowledge”; and Rowley et al., “Building an On-Line Research Community.” 22 For more on the wider issue of Indigenous data sovereignty, see Duarte, Network Sovereignty and Kukutai and Taylor, Indigenous Data Sovereignty. 23 For more on Indigenous classification studies across library, museum, and information studies, see Adler, “The Case for Taxonomic Reparations”; Cherry and Mukunda, “A Case Study in Indigenous Classification”; Lee, “Indigenous Knowledge Organization”; Littletree, “‘Let Me Tell You about Indian Libraries’”; Littletree and Metoyer, “Knowledge Organiza­ tion from an Indigenous Perspective”; Rigby, “Nunavut Libraries Online”; O’Neal, “‘The Right to Know.’” 24 For an excellent example of how one institution supports Indigenous community digital preservation, see the work of the Indigitization Project at the University of British Columbia, Vancouver, BC, at http://indigitization.ca. 25 Jordan, “Documenting and Revitalizing Kiowa Knowledge”; Bell, “A View from the Smith­ sonian”; Bell, “Recovering Voices.” 26 See the SIMA program website, https://naturalhistory.si.edu/research/anthropology/ programs/summer-institute-museum-anthropology.

N o t e s t o pa g e 1 9 2

215

Bibliography

Archival Sources National Anthropological Archives (NAA), Suitland, MD Admin Records, 1972–94 Administrative File 1949–65 Administrative Series File Anthropology, History of, 4903, Diary of Otis T. Mason, 1 July 1884–23 May 1891 Archeology, U.S., 2431: Notes, sketches, and letters used in Smithsonian Annual Reports for 1876–79 and 1881–83 Office Files Otis Tufton Mason Papers, Box 1–5 Photo Lot 4 Photo Lot 70 Series 2 Series 14, Annual Reports, 1920–83 Series 17, Division of Ethnology, Manuscript and Pamphlet File Series 21, Collections Management, 1965–85 Sturtevant Papers, series 1–5 Smithsonian Institution Archives (SIA), Washington, DC RU 000058, Smithsonian Institution, Collected Letters on Ethnology, ca. 1876–79 RU 000065, Smithsonian Institution chief clerk, forms, circulars, announcements, 1846–1933 RU 000120, United States National Museum Office of the Registrar Specimen Records, 1853–1920 RU 000127, USNM Correspondence/Memoranda, 1880–84 RU 000128, USNM Correspondence/Memoranda, 1890–1919 RU 000158, Curator’s Annual Reports RU 007002, Spencer F. Baird Papers, 1833–89 RU 007050, Goode Records

SIA Acc. T89105, United States National Museum, Collected Registers, 1871–1946 SIA Acc. 98-019, Smithsonian Institution Board of Regents, Minutes, 1846–1986

Smithsonian Libraries Smithsonian Institution Annual Reports, 1850–present, various dates National Museum of Natural History (NMNH), Museum Support Center (MSC), Suitland, MD Unpublished Smithsonian staff documents Other Sources Adler, Melissa. “The Case for Taxonomic Reparations.” Knowledge Organization 43, no. 8 (2016): 630–40. Agrawal, Arun. “Indigenous Knowledge and the Politics of Classification.” International Social Science Journal 54, no. 173 (2002): 287–97. –. “The Politics of Indigenous Knowledge.” Australian Academic and Research Libraries 36, no. 2 (2005): 71–81. Anderson, Benedict. Imagined Communities: Reflections on the Origin and Spread of Nation­ alism. New York: Verso, 1991. Anderson, Jane, and Kimberly Christen. “Decolonizing Attribution.” Journal of Radical Lib­ rarianship 5 (2019): 113–52. Appadurai, Arjun. Modernity at Large: Cultural Dimensions of Globalization. Minneapolis: University of Minnesota Press, 1996. Atalay, Sonya. Community-Based Archaeology: Research with, by, and for Indigenous and Local Communities. Berkeley: University of California Press, 2012. –. “No Sense of the Struggle: Creating a Context for Survivance at the NMAI.” American Indian Quarterly 30, no. 5 (2006): 597–618. Baird, Spencer F. Catalogue of North American Birds, Chiefly in the Museum of the Smithsonian Institution. Washington: Smithsonian Institution, 1858. Barad, Karen. Meeting the Universe Halfway: Quantum Physics and the Entanglement of Matter and Meaning. Durham, NC: Duke University Press, 2007. –. “Posthumanist Performativity: Toward an Understanding of How Matter Comes to Matter.” Signs 28, no. 3 (2003): 801–31. Barringer, Tim, and Tom Flynn. Colonialism and the Object: Empire, Material Culture and the Museum. London: Routledge, 1998. Bather, F.A. “Chicago Museum.” In Museums Journal: The Organ of the Museums Association 1. Edited by E. Howarth. London: Dulau, 1902. Bauchspies, Wenda K., and Maria Puig de la Bellacasa. “Feminist Science and Technology Studies: A Patchwork of Moving Subjectivities. An Interview with Geoffrey Bowker, Sandra Harding, Anne Marie Mol, Susan Leigh Star and Banu Subramaniam.” Subjectivity 28, no. 1 (2009): 334–44. Becker, Peter, and William Clark. Little Tools of Knowledge: Historical Essays on Academic and Bureaucratic Practices. Ann Arbor: University of Michigan Press, 2001. Bell, Catherine E., and Val Napoleon, eds. First Nations Cultural Heritage and Law: Case Studies, Voices, and Perspectives. Vancouver: UBC Press, 2008. Bell, Joshua A. “A Bundle of Relations: Collections, Collecting, and Communities.” Annual Review of Anthropology 46 (2017): 241–59. –. “Recovering Voices: Documenting and Sustaining Endangered Languages and Knowledge.” AnthroNotes 32, no. 3 (2011): 7–9.

Bibliography

217

–. “A View from the Smithsonian: Connecting Communities to Collections.” Practicing An­ thro­pology 37, no. 3 (2015): 14–16. Bell, Joshua, Kim Christen, and Mark Turin, “Introduction: After the Return,” Museum An­ thropology Review 7, nos. 1–2 (2013): 1–21. Bennett, Tony. The Birth of the Museum: History, Theory, Politics. London: Routledge, 1995. –. Pasts beyond Memory: Evolution, Museums, Colonialism. London: Routledge, 2004. Bourdieu, Pierre. Outline of a Theory of Practice. Cambridge: Cambridge University Press, 1977. Bowker, Geoffrey C., Karen Baker, Florence Millerand, and David Ribes. “Toward Infor­ mation Infrastructure Studies: Ways of Knowing in a Networked Environment.” In International Handbook of Internet Research, edited by Jeremy Hunsinger, Lisbeth Klastrup, and Matthew Allen, 97–117. Dordrecht: Springer Netherlands, 2009. Bowker, Geoffrey, and Susan Leigh Star. “How Things (Actor-Net) Work: Classification, Magic and the Ubiquity of Standards.” Philosophia 25, nos. 3–4 (1996): 195–220. –. “Invisible Mediators of Action: Classification and the Ubiquity of Standards.” Mind, Culture, and Activity 7, nos. 1–2 (2000): 147–63. –. Sorting Things Out: Classification and Its Consequences. Cambridge, MA: MIT Press, 1999. Bray, Tamara L. Reckoning with the Dead: The Larsen Bay Repatriation and the Smithsonian Institution. Washington, DC: Smithsonian Institution Press, 1994. Brown, Alison K., and Laura Peers, eds. Museums and Source Communities: A Routledge Reader. London: Routledge, 2003. Bruchac, Margaret M. Savage Kin: Indigenous Informants and American Anthropologists. Tucson: University of Arizona Press, 2018. Buettner-Janusch, John. “Boas and Mason: Particularism versus Generalization.” American Anthropologist 59, no. 2 (1957): 318–24. Butler, Judith. Gender Trouble: Feminism and the Subversion of Identity. New York: Routledge, 2006. Callon, Michel. “Some Elements of a Sociology of Translation: Domestication of the Scal­ lops and the Fishermen of St Brieuc Bay.” Sociological Review 32, no. 1 suppl. (1984): 196–233. Campbell, Craig A.R. Agitating Images: Photography against History in Indigenous Siberia. Minneapolis: University of Minnesota Press, 2014. Chari, Sangita, and Jaime M.N. Lavallee. Accomplishing NAGPRA: Perspectives on the Intent, Impact, and Future of the Native American Graves Protection and Repatriation Act. Corvallis: Oregon State University Press, 2013. Charmantier, Isabelle, and Staffan Müller-Wille. “Carl Linnaeus’s Botanical Paper Slips (1767–1773).” Intellectual History Review 24, no. 2 (2014): 215–38. Chenhall, Robert G. Museum Cataloging in the Computer Age. Nashville, TN: American Association for State and Local History, 1975. Chenhall, Robert G., and Peter Homulos. “Propositions for the Future: Museum Data Standards.” Museum International 30, nos. 3–4 (1978): 205–12. Cherry, Alissa, and Keshav Mukunda. “A Case Study in Indigenous Classification: Revisiting and Reviving the Brian Deer Scheme.” Cataloging and Classification Quarterly 53, nos. 5–6 (2015): 548–67. Christen, Kimberly. “Archival Challenges and Digital Solutions in Aboriginal Australia.” SAA Archaeological Record 8, no. 2 (2008): 21–24. Christen, Kimberly, Alex Merrill, and Michael Wynne. “A Community of Relations: Mukurtu Hubs and Spokes.” D-Lib Magazine 23, nos. 5–6 (2017). http://www.dlib.org/dlib/may17/ christen/05christen.html. Christie, Michael. “Words, Ontologies, Aboriginal Databases.” Media International Australia, Incorporating Culture and Policy 116 (August 2005): 52–63.

218

Bibliography

Christie, Michael, and Helen Verran. “Digital Lives in Postcolonial Aboriginal Australia.” Journal of Material Culture 18, no. 3 (2013): 299–317. Clifford, James, and George E. Marcus. Writing Culture: The Poetics and Politics of Ethno­ graphy. Berkeley: University of California Press, 1986. Cobb, Amanda J. “The National Museum of the American Indian as Cultural Sovereignty.” American Quarterly 57, no. 2 (2005): 485–506. Collier-Wise, Kelsey. “Identity Theft: A Search for Legal Protections of Intangible Indigen­ ous Cultural Property.” Great Plains Natural Resources Journal 13 (2010): 85–102. Colwell, Chip. “The Sacred and the Museum: Repatriation and the Trajectories of Inalien­ able Possessions.” Advances in Research – Museum Worlds 2, no. 1 (2014): 10–24. Conaty, Gerald. “The Effects of Repatriation on the Relationship between the Glenbow Museum and the Blackfoot People.” Museum Management and Curatorship 23, no. 3 (2008): 245–59. Conn, Steven. “Heritage vs. History at the National Museum of the American Indian.” Public Historian 28, no. 2 (2006): 69–74. Coote, Jeremy. “Notes and Queries and Social Interrelations: An Aspect of the History of Social Anthropology.” Journal of the Anthropological Society of Oxford 18, no. 3 (1987): 255–72. Cranmer Webster, Gloria. “From Colonization to Repatriation.” In Indigena: Contemporary Native Perspectives, edited by Gerald McMaster and Lee-Ann Martin. Hull, QC: Can­ adian Museum of Civilization, 1992. Cutbill, J.L., A.J. Hallan, and G.D. Lewis. “A Format for the Machine Exchange of Museum Data.” Data Processing in Biology and Geology, edited by J.L. Cutbill. Systematics Association special volume, no. 3 (1971): 255–74. Dagognet, François. Tableaux et langages de la chimie: Essai sur la représentation. 1969; reprint. Seyssel, FR: Champ Vallon, 2002. Darnell, Regna. And Along Came Boas: Continuity and Revolution in Americanist Anthropology. Philadelphia: John Benjamins Publishing, 1998. Daston, Lorraine. “Type Specimens and Scientific Memory.” Critical Inquiry 31, no. 1 (2004): 153–82. Daston, Lorraine, and Peter Galison. Objectivity. New York: Zone Books, 2007. Dauber, Kenneth. “Bureaucratizing the Ethnographer’s Magic.” Current Anthropology 36, no. 1 (1995): 75–95. DeHass, Medeia Csoba, and Eric Hollinger. “3D Heritage Preservation and Indigenous Communities in the Circumpolar North.” Arctic Yearbook (2018): 1–15. Dias, Nélia. “The Visibility of Difference: Nineteenth-Century French Anthropological Col­lections.” In The Politics of Display: Museums, Science, Culture, edited by Sharon Macdonald, 36–52. London: Routledge, 1998. Douglas, Mary. Purity and Danger: An Analysis of Concepts of Pollution and Taboo. Harmonds­ worth, UK: Penguin, 1970. Doyle, Ann. “Naming, Claiming, and (Re)Creating: Indigenous Knowledge Organization at the Cultural Interface.” PhD diss., University of British Columbia, 2013. Duarte, Marisa Elena. Network Sovereignty: Building the Internet across Indian Country. Seattle: University of Washington Press, 2017. Duarte, Marisa Elena, and Miranda Belarde-Lewis. “Imagining: Creating Spaces for In­ digenous Ontologies.” Cataloging and Classification Quarterly 53, nos. 5–6 (2015): 677–702. Echo-Hawk, Roger C. Keepers of Culture: Repatriating Cultural Items under the Native Amer­ ican Graves Protection and Repatriation Act. Denver, CO: Denver Art Museum, 2002. Edwards, Elizabeth, and Janice Hart. Photographs Objects Histories: On the Materiality of Images. London: Routledge, 2004.

Bibliography

219

England, Kim, and Kate Boyer. “Women’s Work: The Feminization and Shifting Meanings of Clerical Work.” Journal of Social History 43, 2 (2009): 307–40. Evans, Jessica, and David Boswell, eds. Representing the Nation – A Reader: Histories, Heritage and Museums. London: Routledge, 1999. Fenton, William N. “The Museum and Anthropological Research.” Curator: The Museum Journal 3, no. 4 (1960): 327–55. Fforde, Cressida, Jane Hubert, and Paul Turnbull. The Dead and Their Possessions: Repatriation in Principle, Policy and Practice. London: Routledge, 2004. Fleck, Ludwik. Genesis and Development of a Scientific Fact. Chicago: University of Chicago Press, 1979. Foucault, Michel. Discipline and Punish: The Birth of the Prison. 2nd ed. New York: Vintage Books, 1995. –. The Order of Things: An Archaeology of the Human Sciences. London: Routledge, 2002. Fujimura, Joan. “Authorizing Knowledge in Science and Anthropology.” American Anthro­ pologist 100, no. 2 (1998): 347–60. Geismar, Haidy. Museum Object Lessons for the Digital Age. London: UCL Press, 2018. Geismar, Haidy, and William Mohns. “Social Relationships and Digital Relationships: Re­ thinking the Database at the Vanuatu Cultural Centre.” Journal of the Royal Anthropological Institute 17 (2011): S133–S155. Genoways, Hugh H., and Mary Anne Andrei. Museum Origins: Readings in Early Museum History and Philosophy. Walnut Creek, CA: Left Coast Press, 2008. Gibson, Laura Kate. “Decolonising South African Museums in a Digital Age: Re-imagining the Iziko Museums’ Natal Nguni Catalogue and Collection,” PhD diss., Kings College London, 2019. Gitelman, Lisa. Paper Knowledge: Toward a Media History of Documents. Durham, NC: Duke University Press, 2014. Gitelman, Lisa, and Virginia Jackson, “Introduction.” In “Raw Data” Is an Oxymoron, edited by Lisa Gitelman, 1–14. Cambridge, MA: MIT Press, 2013. Glass, Aaron. “Drawing on Museums: Early Visual Fieldnotes by Franz Boas and the Indigen­ ous Recuperation of the Archive.” American Anthropologist 120, no. 1 (2018): 72–88. Greene, Candace. “Documentation, Attribution and the Ideal Type.” In Art and Artifacts: Essays in Material Culture and Museum Studies in Honor of Jane Powell Dwyer, edited by Harold David Juli, 9–18. Department of Anthropology, Brown University, 1992. –. “Material Connections: ‘The Smithsonian Effect’ in Anthropological Cataloguing.” Museum Anthropology 39, no. 2 (2016): 147–62. Griaule, Marcel. Méthode de l’ethnographie. Paris: Presses Universitaires de France, 1957. Griffiths, Alison. Wondrous Difference: Cinema, Anthropology, and Turn-of-the-Century Visual Culture. New York: Columbia University Press, 2002. Haas, Angela M. “Wampum as Hypertext: An American Indian Intellectual Tradition of Multimedia Theory and Practice.” Studies in American Indian Literatures 19, no. 4 (2008): 77–100. Halpern, Orit. Beautiful Data: A History of Vision and Reason since 1945. Durham, NC: Duke University Press, 2014. Haraway, Donna. “A Manifesto for Cyborgs: Science, Technology, and Socialist Feminism in the 1980s.” Socialist Review 80 (1985): 65–108. –. “Teddy Bear Patriarchy: Taxidermy in the Garden of Eden, New York City, 1908–1936.” Social Text 11 (1984): 20–64. Harding, Sandra, ed. The Feminist Standpoint Theory Reader: Intellectual and Political Controversies. London: Routledge, 2003. Harding, Sandra. Sciences from Below: Feminisms, Postcolonialities, and Modernities. Durham, NC: Duke University Press, 2008.

220

Bibliography

Hays-Gilpin, Kelley, and Ramson Lomatewama. “Curating Communities at the Museum of Northern Arizona.” In Reassembling the Collection: Ethnographic Museums and Indigenous Agency, edited by Sarah Byrne and Rodney Harrison, 259–84. Santa Fe, NM: SAR Press, 2013. Hetherington, Kevin. “From Blindness to Blindness: Museums, Heterogeneity and the Subject.” Sociological Review 47, no. 1 suppl. (1999): 51–73. Hinsley, Curtis M. Savages and Scientists: The Smithsonian Institution and the Development of American Anthropology, 1846–1910. Washington, DC: Smithsonian Institution Press, 1981. –. The Smithsonian and the American Indian: Making a Moral Anthropology in Victorian America. Washington, DC: Smithsonian Institution Press, 1994. Hochman, Brian. Savage Preservation: The Ethnographic Origins of Modern Media Technology. Minneapolis: University of Minnesota Press, 2014. Hollinger, R. Eric. Assessment of Tlingit Objects Requested by the Hoonah Indian Association for Repatriation as Unassociated Funerary Objects in the Collections of the National Museum of Natural History, Smithsonian Institution. Washington, DC: Repatriation Office, National Museum of Natural History, 2012. Hollinger, R. Eric, Edwell John Jr., Harold Jacobs, Lora Moran-Collins, Carolyn Thome, Jonathan Zastrow, Adam Metallo, Günter Waibel, and Vince Rossi. “Tlingit-Smithsonian Collaborations with 3D Digitization of Cultural Objects.” Museum Anthropology Review 7, nos. 1–2 (2013): 201–53. Holmes, William Henry, and Otis T. Mason. “Instructions to Collectors of Historical and Anthropological Specimens: Especially Designed for Collectors in the Insular Possessions of the United States.” Part Q of Bulletin of the United States National Museum, no. 39 (1902): 1–16. Hooper-Greenhill, Eilean. Museums and the Shaping of Knowledge. London: Routledge, 1992. Hull, Matthew S. “Documents and Bureaucracy.” Annual Review of Anthropology 41, no. 1 (2012): 251–67. Hunt, Sarah. “Ontologies of Indigeneity: The Politics of Embodying a Concept.” Cultural Geographies 21, no. 1 (2014): 27–32. Isaac, Gwyneira. “What Are Our Expectations Telling Us? Encounters with the NMAI.” American Indian Quarterly 30, no. 3 (2006): 574–96. –. “Whose Idea Was This?” Current Anthropology 52, no. 2 (2011): 211–33. Isaac, Gwyneira, and Barbara Isaac. “Uncovering the Demographics of Collecting: A CaseStudy of the US Exploring Expedition (1838–1842).” Journal of the History of Collections 28, no. 2 (2015): 209–23. Isaac, Jaimie Lyn. “Decolonizing Curatorial Practice: Acknowledging Indigenous Curatorial Praxis, Mapping Its Agency, Recognizing Its Aesthetic within Contemporary Canadian Art.” Master’s thesis, University of British Columbia, 2016. Jacknis, Ira. “A New Thing? The NMAI in Historical and Institutional Perspective.” American Indian Quarterly 30, nos. 3–4 (2006): 511–42. Jenkins, David. “Object Lessons and Ethnographic Displays: Museum Exhibitions and the Making of American Anthropology.” Comparative Studies in Society and History 36, no. 2 (1994): 242–70. Jordan, Michael Paul. “Documenting and Revitalizing Kiowa Knowledge: Material Culture Studies and Community Engagement.” Museum Anthropology Review 9, nos. 1–2 (2015): 80–95. Judd, Neil Merton. The Bureau of American Ethnology: A Partial History. Norman: University of Oklahoma Press, 1967. Kittler, Friedrich A. Gramophone, Film, Typewriter. Redwood City, CA: Stanford University Press, 1999.

Bibliography

221

Knell, Simon J. “Museums, Reality and the Material World.” In Museums in the Material World, edited by Simon Knell, 1–28. London: Routledge, 2007. Knight, Emma. “Unpacking the Museum Register.” Museum Worlds 5, no. 1 (2017): 35–47. Knorr-Cetina, Karin. The Manufacture of Knowledge. Oxford: Pergamon Press, 1981. Kohlstedt, Sally. “Otis T. Mason’s Tour of Europe: Observation, Exchange, and Stan­dard­ ization in Public Museums, 1889.” Museum History Journal 1, no. 2 (2008): 181–208. Krajewski, Markus. Paper Machines: About Cards and Catalogs, 1548–1929. Cambridge, MA: MIT Press, 2011. Krmpotich, Cara. The Force of Family: Repatriation, Kinship, and Memory on Haida Gwaii. Toronto: University of Toronto Press, 2014. Krmpotich, Cara, and Laura Peers. This Is Our Life: Haida Material Heritage and Changing Museum Practice. Vancouver: UBC Press, 2013. Kuhn, Thomas S. The Structure of Scientific Revolutions. 4th edition. Chicago: University of Chicago Press, 2012. Kukutai, Tahu, and John Taylor, eds. Indigenous Data Sovereignty: Toward an Agenda. Can­ berra: ANU Press, 2016. Lampland, Martha, and Susan Leigh Star. Standards and Their Stories: How Quantifying, Classifying, and Formalizing Practices Shape Everyday Life. Ithaca, NY: Cornell University Press, 2009. Landow, George P. HyperText: The Convergence of Contemporary Critical Theory and Technol­ ogy. Baltimore, MD: Johns Hopkins University Press, 1991. Latham, Kiersten F. “Museum Object as Document: Using Buckland’s Information Concepts to Understand Museum Experiences.” Journal of Documentation 68, no. 1 (2012): 45–71. Latour, Bruno. Laboratory Life: The Construction of Scientific Facts. Princeton, NJ: Princeton University Press, 1986. –. Pandora’s Hope: Essays on the Reality of Science Studies. Cambridge, MA: Harvard University Press, 1999. –. Reassembling the Social: An Introduction to Actor-Network-Theory. Oxford: Oxford Uni­ versity Press, 2007. –. Science in Action: How to Follow Scientists and Engineers through Society. Cambridge, MA: Harvard University Press, 1987. –. “Visualization and Cognition: Thinking with Eyes and Hands.” Knowledge and Society 6 (1986): 1–40. –. “Where Are the Missing Masses? The Sociology of a Few Mundane Artifacts.” In Shaping Technology / Building Society: Studies in Sociotechnical Change, edited by Wiebe Bijker and John Law, 225–58. Cambridge, MA: MIT Press, 1992. Latour, Bruno, and Steve Woolgar. Laboratory Life: The Construction of Scientific Facts. Princeton, NJ: Princeton University Press, 1986. Law, John. “Actor Network Theory and Material Semiotics.” The New Blackwell Companion to Social Theory, edited by Brian S. Turner, 141–58. Oxford: Wiley-Blackwell, 2009. –. “Notes on the Theory of the Actor-Network: Ordering, Strategy, and Heterogeneity.” Systemic Practice and Action Research 5, no. 4 (1992): 379–95. –. On Power and Its Tactics: A View from the Sociology of Science.” Sociological Review 34, 1 (1986): 1–38. Lee, Deborah. “Indigenous Knowledge Organization: A Study of Concepts, Terminology, Structure and (Mostly) Indigenous Voices.” Partnership: The Canadian Journal of Library and Information Practice and Research 6, no. 1 (2011): 1–33. Lemov, Rebecca. Database of Dreams: The Lost Quest to Catalog Humanity. New Haven, CT: Yale University Press, 2015.

222

Bibliography

Leroi-Gourhan, André, Évolution et techniques I: L’homme et la matière. 2nd ed. Paris: Albin Michel, 1971. Littletree, Sandra. “‘Let Me Tell You about Indian Libraries’: Self-Determination, Leadership, and Vision – The Basis of Tribal Library Development in the United States.” PhD diss., University of Washington, 2018. Littletree, Sandra, and Cheryl A. Metoyer. “Knowledge Organization from an Indigenous Perspective: The Mashantucket Pequot Thesaurus of American Indian Terminology Project.” Cataloging and Classification Quarterly 53, nos. 5–6 (2015): 640–57. Lonetree, Amy. Decolonizing Museums: Representing Native America in National and Tribal Museums. Chapel Hill: University of North Carolina Press, 2012. –. “Missed Opportunities: Reflections on the NMAI.” American Indian Quarterly 30, nos. 3–4 (2006): 632–45. Lubar, Steven. Inside the Lost Museum: Curating, Past and Present. Cambridge, MA: Harvard University Press, 2017. Lyons, Natasha, David M. Schaepe, Kate Hennessy, Michael Blake, Clarence Pennier, John R. Welch, Kyle McIntosh, et al. “Sharing Deep History as Digital Knowledge: An Ontology of the Sq’éwlets Website Project.” Journal of Social Archaeology 16, no. 3 (2016): 359–84. Mason, Otis T. Aboriginal American Basketry: Studies in a Textile Art without Machinery. Washington, DC: Government Printing Office, 1904. –. Ethnological Directions Relative to the Indian Tribes of the United States. Washington, DC: Smithsonian Institution, 1875. –. “The Occurrence of Similar Inventions in Areas Widely Apart.” Science 226 (1887): 534–35. –. The Origins of Invention: A Study of Industry among Primitives Peoples. Washington, DC: Government Printing Office, 1895. Mattern, Shannon. Code and Clay, Data and Dirt: Five Thousand Years of Urban Media. Min­neapolis: University of Minnesota Press, 2017. McFeat, Tom. “The Object of Research in Museums.” Contributions to Ethnology 5, Bulletin 204: 91–99. Ottawa: National Museums of Canada, 1967. McKenzie, Donald F. “The Sociology of a Text: Orality, Literacy and Print in Early New Zealand.” Library 6, no. 4 (1984): 333–65. Merrill, William L. The Return of the Ahayu:Da: Lessons for Repatriation from Zuni Pueblo and the Smithsonian Institution. Chicago: University of Chicago Press, 1993. –. “William Curtis Sturtevant, Anthropologist.” In Anthropology, History, and American Indians: Essays in Honor of William Curtis Sturtevant, edited by William L. Merrill and Ives Goddard, 11–36. Washington, DC: Smithsonian Institution Press, 2002. Meyer, Adolf Bernhard. Studies of the Museums and Kindred Institutions of New York City, Albany, Buffalo, and Chicago. Vol. 133. Washington, DC: US Government Printing Office, 1905. Mol, Annemarie. The Body Multiple: Ontology in Medical Practice. Durham, NC: Duke Uni­ versity Press, 2002. Moreton-Robinson, Aileen. The White Possessive: Property, Power, and Indigenous Sovereignty. Minneapolis: University of Minnesota Press, 2015. Moser, Stephanie. “Making Expert Knowledge through the Image: Connections between Antiquarian and Early Modern Scientific Illustration.” Isis 105, no. 1 (2014): 58–99. Muhr, Sara Louise, and Alf Rehn. “On Gendered Technologies and Cyborg Writing.” Gender, Work and Organization 22, no. 2 (2015): 129–38. Muntean, Reese, Kate Hennessy, Alissa Antle, Susan Rowley, Jordan Wilson, and Brendan Matkin. “ʔeləw ̓ k ̓ ʷ – Belongings: Tangible Interactions with Intangible Heritage.” Journal of Science and Technology of the Arts 7, no. 2 (2015): 59–69.

Bibliography

223

Muntean, Reese, Hannah Turner, and Kate Hennessy. “Addressing Digital Affordance: Mapping Cross Cultural Value in 3D Scanned Belongings.” Paper presented at the Art, Materiality and Representation Conference, Royal Anthropological Institute, London, June 2018. Nakata, Martin, Victoria Nakata, Sarah Keech, and Reuben Bolt. “Decolonial Goals and Pedagogies for Indigenous Studies.” Decolonization: Indigeneity, Education and Society 1, no. 1 (2012): 120–40. Nichols, Catherine. “A Century of Circulation: The Return of the Smithsonian Institution’s Duplicate Anthropological Specimens.” Museum Anthropology 37, no. 2 (2014): 144–59. –. “Lost in Museums: The Ethical Dimensions of Historical Practices of Anthropological Specimen Exchange.” Curator: The Museum Journal 57, no. 2 (2014): 225–36. –. “Museum Networks: The Exchange of the Smithsonian Institution’s Duplicate Anthro­ pology Collections.” PhD diss., Arizona State University, 2014. Nichols, Catherine A., and Nancy J. Parezo. “Social and Material Connections: Otis T. Mason’s European Grand Tour and Collections Exchanges.” History and Anthropology 28, no. 1 (2016): 58–83. Nikolova-Houston, Tatiana. “Margins and Marginality: Marginalia and Colophons in South Slavic Manuscripts during the Ottoman Period, 1393–1878.” PhD diss., University of Texas at Austin, 2008. O’Neal, Jennifer. “‘The Right to Know’: Decolonizing Native American Archives.” Journal of Western Archives 6, no. 1 (2015): P1–17. Opp, James. “The Colonial Legacies of the Digital Archive: The Arnold Lupson Photographic Collection.” Archivaria 65 (2008): 3–19. Parezo, Nancy J. Anthropology Goes to the Fair: The 1904 Louisiana Purchase Exposition. Lincoln: University of Nebraska Press, 2007. –. “Cushing as Part of the Team: The Collecting Activities of the Smithsonian Institution.” American Ethnologist 12, no. 4 (1985): 763–74. –. “The Formation of Ethnographic Collections: The Smithsonian Institution in the Amer­ ican Southwest.” Advances in Archaeological Method and Theory 10 (1 January 1987): 1–47. Parks, Lisa. “‘Stuff You Can Kick’: Toward a Theory of Media Infrastructures.” In Between Humanities and the Digital, edited by Patrik Svensson and David Theo Goldberg, 355–73. Cambridge, MA: MIT Press, 2015. Parry, Ross. Recoding the Museum: Digital Heritage and the Technologies of Change. London: Routledge, 2007. Pearce, Susan. On Collecting: An Investigation into Collecting in the European Tradition. London: Routledge, 1995. Petch, Alison. “Notes and Queries and the Pitt Rivers Museum.” Museum Anthropology 30, no. 1 (2007): 21–39. Phillips, Ruth B. Museum Pieces: Toward the Indigenization of Canadian Museums. Vol. 6. Montreal and Kingston: McGill-Queen’s University Press, 2011. Poovey, Mary. A History of the Modern Fact: Problems of Knowledge in the Sciences of Wealth and Society. Chicago: University of Chicago Press, 1998. Povinelli, Elizabeth A. “The Woman on the Other Side of the Wall: Archiving the Other­ wise in Postcolonial Digital Archives.” Differences 22, no. 1 (2011): 146–71. Powell, John Wesley. Introduction to the Study of Indian Languages: With Words, Phrases, and Sentences to Be Collected. Washington, DC: Government Printing Office, 1877. Pratt, Mary Louise. Imperial Eyes: Travel Writing and Transculturation. London: Routledge, 1992. Rau, Charles. “Circular Relative to the Contribution of Aboriginal Antiquities to the United States National Museum.” Proceedings of the United States National Museum, no. 21 (1883).

224

Bibliography

Reardon, Jenny, and Kim TallBear. “‘Your DNA Is Our History’: Genomics, Anthropology, and the Construction of Whiteness as Property.” Current Anthropology 53, no. S5 (2012): S233–S245. Reid, Kirsty, and Fiona Paisley, eds. Critical Perspectives on Colonialism: Writing the Empire from Below. New York: Routledge, 2013. Ricciardelli, Alex F. A Pilot Study for Inventorying Ethnological Collections. Norman: Stovall Museum of Science and History and the University of Oklahoma, 1967. Rigby, Carol. “Nunavut Libraries Online Establish Inuit Language Bibliographic Catalo­ ging Standards: Promoting Indigenous Language Using a Commercial ILS.” Cataloging and Classification Quarterly 53, nos. 5–6 (2015): 615–39. Riles, Annelise. Documents: Artifacts of Modern Knowledge. Ann Arbor: University of Michi­ gan Press, 2006. Rivinus, E.F., and E.M. Youssef. Spencer Baird of the Smithsonian. Washington, DC: Smith­ sonian Institution Press, 1992. Robbins, Helen A. 2014. “Historic Challenges, Relationships and the Future: A Long and Crooked Road to Hopi Haven.” Paper presented at the 113th meeting of the American Anthropological Association, Washington, DC. 4 December 2014. Roberts, D. Andrew, and Richard B. Light. “Progress in Documentation: Museum Docu­ mentation.” Journal of Documentation 36, no. 1 (1980): 42–84. Roberts, Frank H.H. “One Hundred Years of Smithsonian Anthropology.” Science 104, no. 2693 (1946): 119–25. Rosenberg, Daniel. “Data before the Fact.” In “Raw Data” Is an Oxymoron, edited by Lisa Gitelman, 15–40. Cambridge, MA: MIT Press, 2013. Rosoff, Nancy B. “Integrating Native Views into Museum Procedures: Hope and Practice at the National Museum of the American Indian.” In Museums and Source Communities: A Routledge Reader, edited by Alison K. Brown and Laura Peers, 72–79. London: Rout­ ledge, 2003. Rowley, Susan. “The Reciprocal Research Network: The Development Process.” Museum Anthropology Review 7, nos. 1–2 (2013): 22–43. Rowley, Susan, Dave Schaepe, Leona Sparrow, Andrea Sanborn, Ulrike Radermacher, Ryan Wallace, Nicholas Jakobsen, Hannah Turner, Sivia Sadofsky, and Tristan Goffman. “Build­ ing an On-Line Research Community: The Reciprocal Research Network.” In Museums and the Web, 2010: Proceedings. Toronto: Archives and Museum Informatics, 2010. Rydell, Robert W. All the World’s a Fair: Visions of Empire at American International Expositions, 1876–1916. Chicago: University of Chicago Press, 2013. Sarr, Felwine, and Bénédicte Savoy. The Restitution of African Cultural Heritage: Toward a New Relational Ethics. November 2018. http://restitutionreport2018.com/. Simpson, Moira G. Making Representations: Museums in the Post-Colonial Era. London: Routledge, 1996. Smith, Linda Tuhiwai. Decolonizing Methodologies: Research and Indigenous Peoples. London: Zed Books; University of Otago Press, 1999. Star, Susan Leigh. “The Ethnography of Infrastructure.” American Behavioral Scientist 43, no. 3 (1999): 377–91. Star, Susan Leigh, and Karen Ruhleder. “Steps toward an Ecology of Infrastructure: Design and Access for Large Information Spaces.” Information Systems Research 7, no. 1 (1996): 111–34. Sterne, Jonathan. MP3: The Meaning of a Format. Durham, NC: Duke University Press, 2012. Stoler, Ann. Along the Archival Grain: Epistemic Anxieties and Colonial Common Sense. Princeton, NJ: Princeton University Press, 2010.

Bibliography

225

–. “Colonial Archives and the Arts of Governance.” Archival Science 2, nos. 1–2 (2002): 87–109. –. Duress: Imperial Durabilities in Our Times. Durham, NC: Duke University Press, 2016. Strathern, Marilyn. Partial Connections. Sabage, MD: Rowman and Littlefield, 1991. Sturtevant, William C. Guide to Field Collecting of Ethnographic Specimens. Washington, DC: Smithsonian Institution, 1969. –. Handbook of North American Indians: California. Washington, DC: Government Printing Office, 1978. –. “Studies in Ethnoscience.” American Anthropologist 66, no. 3 (1964): 99–131. Szymanski, Margaret H., and Jack Whalen. Making Work Visible: Ethnographically Grounded Case Studies of Work Practice. Cambridge: Cambridge University Press, 2011. TallBear, Kimberly. Native American DNA: Tribal Belonging and the False Promise of Genetic Science. Minneapolis: University of Minnesota Press, 2013. Thomas, Cyrus. “Directions for Mound Exploration.” Circular No. 28. Proceedings of the United States National Museum 4, no. 28 (1884): 581–83. Thomas, Jennifer M. “The Documentation of the British Museum’s Natural History Col­ lections, 1760–1836.” Archives of Natural History 39, no. 1 (2012): 111–25. Todd, Zoe. “An Indigenous Feminist’s Take on the Ontological Turn: ‘Ontology’ Is Just Another Word for Colonialism.” Journal of Historical Sociology 29, no. 1 (2016): 4–22. Trigger, Bruce. A History of Archaeological Thought. Cambridge: Cambridge University Press, 1989. Trope, Jack F., and Walter R. Echo-Hawk. “The Native American Graves Protection and Repatriation Act: Background and Legislative History.” In Repatriation Reader: Who Owns American Indian Remains, edited by Devon A. Mihesuah, 123–68, Lincoln: University of Nebraska Press, 2000. Tuck, Eve, and K. Wayne Yang. “Decolonization Is Not a Metaphor.” Decolonization: In­ digeneity, Education and Society 1, no. 1 (2012). Turner, Hannah. “The Computerization of Material Culture Catalogues: Objects and Infra­ structure in the Smithsonian Institution’s Department of Anthropology.” Museum Anthro­ pology 39, no. 2 (2016): 163–77. –. “Critical Histories of Museum Catalogues.” Museum Anthropology 39, no. 2 (2016): 102–10. --. “Decolonizing Ethnographic Documentation: A Critical History of the Early Museum Catalogs at the Smithsonian’s National Museum of Natural History.” Cataloging and Classification Quarterly 53, no. 5–6 (2015): 658–76. –. “Organizing Knowledge in Museums: A Review of Concepts and Concerns.” Knowledge Organization 44, no. 7 (2017). Turner, Hannah, and Candace Greene. “Access to Native Collections in Museums and Archives: History, Context, and Future Directions.” In Handbook of North American Indians (HNAI), vol. 1. Smithsonian Institution Scholarly Press, forthcoming. Turner, Hannah, Kate Hennessy, Meghann O’Brien, and Conrad Sly. “New Media: ‘Wrapped in the Cloud’: An Interview with Meghann O’Brien and Conrad Sly.” BC Studies 200 (2019): 125–40. Urban, Richard J. “Library Influence on Museum Information Work.” Library Trends 62, no. 3 (2014): 596–612. Urry, James. “‘Notes and Queries on Anthropology’ and the Development of Field Methods in British Anthropology, 1870–1920.” Proceedings of the Royal Anthropological Institute of Great Britain and Ireland, no. 1972 (1972): 45–57. Van Allen, Adrian. “Pinning Beetles, Biobanking Futures: Practices of Archiving Life in a Time of Extinction,” New Genetics and Society 37, no. 4 (2018): 387–410.

226

Bibliography

Vance, David. “Museum Computer Network: Progress Report.” Museologist no. 135 (1975): 3–10. Verdon, Michel. “The World Upside Down: Boas, History, Evolutionism, and Science.” History and Anthropology 17, no. 3 (2006): 171–87. Vizenor, Gerald. Survivance: Narratives of Native Presence. Lincoln: University of Nebraska Press, 2008. Vygotsky, Lev Semenovich. Language and Thought. Cambridge, MA: MIT Press, 1962. Walklate, Jennifer. “The Quotidian and the Bizarre: Rewriting the Day Book.” Master’s thesis, University of Leicester, 2009. Walsh, Jane MacLaren. “Collections as Currency.” Smithsonian Contributions to Anthropology 44 (2002): 201–9. Waterton, Claire. “Experimenting with the Archive: STS-Ers as Analysts and Co-Constructors of Databases and Other Archival Forms.” Science, Technology and Human Values 35, no. 5 (2010): 645–76. Watkins, Joe. “Becoming American or Becoming Indian? NAGPRA, Kennewick and Cultural Affiliation.” Journal of Social Archaeology 4, no. 1 (2004): 60–80. Weld, Kirsten. Paper Cadavers: The Archives of Dictatorship in Guatemala. Durham, NC: Duke University Press, 2014. Wemigwans, Jennifer. A Digital Bundle: Protecting and Promoting Indigenous Knowledge Online. Regina: University of Regina Press, 2018. Wilson, Jordan. “Belongings’ in Cə̓ snaʔəm: the City before the City.” IPinCH: Intellectual Property Issues in Cultural History. Posted 27 January 2016. http://www.sfu.ca/ipinch/ outputs/blog/citybeforecitybelongings/ Wright, Alex. Cataloging the World: Paul Otlet and the Birth of the Information Age. Oxford: Oxford University Press, 2014. Wylie, Alison. “Why Standpoint Matters.” In Science and Other Cultures: Issues in Philosophies of Science and Technology, edited by Robert Figueroa and Sandra G. Harding, 26–48. London: Routledge, 2003.

Bibliography

227

Index

Note: For abbreviations used in subheadings, refer to the list of abbreviations on page xiv. “(f )” after a page number indicates a figure and “(t)” a table. Abbot, Ezra, 90 access to collections, 174–75, 179–81, 190, 213n53 accession files, 111–12, 147, 151, 190, 214n12 accession number (data field), 73, 87–88, 94, 117, 213n41 accessions: cataloguing catches up to rate of, 115–16; defined, 67, 104; history, 65, 202n1; process, 112, 115, 134–35, 205n29 administrators, 92 ADP (Automatic Data Processing) unit, 138, 139(f )–40(f ), 142 African collections, 148 Alaska, 23, 43, 69, 173, 176. See also Hoonah Repatriation Collection American Association of Museums, 104, 106 American Ethnological Society, 38 American Museum of Natural History, 68, 196n36, 199n29 American museums, other: card catalogues in, 90; Field Museum of Natural History, 68–69, 104; Peabody Museum, 13, 68, 70, 204n47 animal specimens. See desiderata; natural history specimens announcement books, 92

annual reports, Smithsonian’s: on accessioning process, 205n29; on aid provided for expeditions, 28; archaeology code system in, 201n92; archive of, 107; on cataloguing and accessions, 115–16; cataloguing given own section in, 103; on cataloguing process, 104, 105; on circulars, 37, 38, 47–48, 52, 199n37; on classification, 97, 205n31, 206n70; on data entry, 167; on data’s value, 59–60; on development of early ledger books, 70–71; on errors in card catalogues, 114– 15; on Hoonah Repatriation Collection, 79; on institution’s financial problems, 110; on object exchanges and duplicates, 206n63; on object’s journey from field to storage location, 92; on parameters for curators’ work, 95; on tribal affiliation, 198n20; on tribal index, 112 anthropology: approach to, 7; changing methods of, 111, 126–27, 133–34; differing needs of collections in archaeology vs, 165–66; history of colonialism as history of, 13; physical, 104, 180, 213n53; professionalization of, 57; as science, 8, 33. See also Department of Anthropology; ethnology

“Anthropology Data Manual,” 160, 162 Anthropology Inventory Worksheet, 144–46 antiquities, 40, 45 archaeology: circulars on, 39, 42, 43–44, 46, 50–53, 200n60, 201n96; code systems for, 201n92; data excluded from online database, 180–81; differing needs of collections in anthropology vs, 165– 66; evolutionary approach to, 196n34; material culture’s increasing association with, 126; object drawing in, 76–77; vs ethnology, 44 archives, 25–26, 107, 212n14 Army Medical Museum, 104 Assessment of Tlingit Objects (Hollinger), 203n43 assistants and aides: Foreman, 46–47, 75, 76(f ); routine work of, 111, 120, 136; work on announcement books, 92; work on card catalogues, 95, 103, 111, 120; work on Hoonah Repatriation Collection, 79 audio media, 14 Audubon, John, 202n8 Australia, 171 authentication, 41, 95, 116 authority files, 162 authority name (data field), 177 Automatic Data Processing (ADP) unit, 138, 139(f )–40(f ), 142 Ayers, Ed. E., 69 backlogs, 111, 115, 116, 119, 162 bad data, 20–21, 167–72, 182. See also errors and inconsistencies BAE. See Bureau of American Ethnology (BAE) Baird, Spencer F.: appointments at Smithsonian, 31, 46–47, 70, 198n22; Audubon recommends, 202n8; circulars by, 34–38, 43–44, 46; Foreman and, 75, 76; influence on classification, 85; Mason’s letters to, 94, 96; on object’s journey from field to storage location, 92; role in establishing ledger books, 69–71, 73 baskets, 205n26 Belarde-Lewis, Miranda, 6 “belongings” (as term), 195n9

Bender, Clarence, 120 Bennett, Tony, 10, 11, 12, 29, 93 Bhaktin, Mikhail, 77 bibliographies, 89 Bibliotheca Universalis (Gessner), 89 “Black South African” (as term), 171 “blanks” (as term), 197n69 Boas, Franz, 84, 92 Bolles, Timothy Dixon, 18, 78–82, 84, 203n38, 203n43, 204n46 books and preservation, 65 Bourdieu, Pierre, 197n64 Bowker, Geoffrey, 14, 75, 157 boxes. See data fields Britain and United Kingdom, 30, 104, 141, 200n71 British Museum, 98–99. See also Pitt Rivers Museum Bruchac, Margaret, 7, 26, 59, 184 Bureau of American Ethnology (BAE): abolished, 128, 200n76; created, 9, 31, 47; objects enter collections through, 116; Sturtevant’s work at, 127; USNM’s increasing focus on, 105 Bureau of Indian Affairs, 31, 46, 48 cabinets of curiosity, vii, 11 Callon, Michel, 15 Canadian Institute, 53 card catalogues, 87–124 —1930s to 1960s reworking of, 111–16, 118, 119–20, 122–23 —accessible in Museum Support Center, 25 —approach to, 14, 19–20, 88, 90–91, 123, 187 —criticism of, 103, 110, 117, 118, 135 —digitization of: basis of early computerized records, 126, 151; difficulty of, 150–51, 164, 166–67, 168–69; early computerized records complement catalogues, 142; images in EMu, 160, 180, 181; non-specialists’ work on, 143; purpose, 141, 182; underlining and, 164, 165(f) —early use in Europe, 89 —as foundational media technology, 5 —future research on, 191–92 —importance of access to, 190 —influence of repatriation legislation on, 175

Index

229

—as instrument of organizing chaos, 91–95 —Jewett’s vision of, 90, 204n8 —Latham’s work on, 204n21 —ledger books used with, 69, 102–3, 115, 206n63 —Mason credited with adoption of, 86, 88–89 —Mason’s vision of, 19, 91, 93, 96–97, 98–99, 102, 104–5, 112 —as means of producing knowledge, 157 —naming issues in, 99, 170–71 —object name field in, 150–51, 155, 163–64, 166 —photos, 118(f ), 119(f ), 121(f ) —praise for, 104 —processing of entering data into, 102, 105 —reconciled with inventory index, 147 —routine work of, 105–11, 120 —similitude and, 82, 98 —types of staff working on, 103 —use of inventory to correct errors in, 142, 144, 147 —value of, 101–2, 105, 118 (see also researchers: card catalogues designed to benefit) —vs monographic catalogues, 205n47 —vs museum catalogues, 204n24 catalogue books. See ledger books catalogue number (data field): in card catalogues, 111, 113, 114, 117; in databases, 160–61; in ledger books, 71; as organizing principle for collections, 87–88; printed on specimens, 115; used in search for errors, 168 “Catalogue of North American Birds” (Baird), 69 cataloguers: and amateur ethnologists, 13; approach to, 185; frustrations with early computerized records, 167; names of, 161; objects’ shifting locality and, 175– 76; reconfigure naming conventions, 168; struggle to classify Indigenous materials, 6. See also clerks catalogues, Bolles’s, 79–81 Cataloguing the World (Wright), 87 categorization. See classification Ceco’q (Tlingit word), 83, 203n43 Centennial Exposition collections, 31, 48 Chenhall, Robert, 210n42 Cherokee materials, 115

230

Index

Chicago museums, 68–69, 104 Chinook Jargon, 172 Churchill, S., 37 “Circular” (Canadian Institute, 1889), 53 Circular no. 1 (1852), 38, 62, 199n6 Circular no. 316 (Henry, 1878), 51–53 “Circular Relating to Collections in Archeology and Ethnology” (Henry, 1867), 43–44, 46, 200n60 “Circular Relative to Contribution of Aboriginal Antiquities to the United States National Museum” (Rau, 1883), 53, 201n96 circulars and field guides, 28–64; approach to, 17–18, 30, 186; Baird’s, 34–38, 46; British, 30, 200n71; and data gaps, 70; first mention in annual reports, 199n37; as foundational media technology, 5; future research on, 191–92; general directions provided in, 34; Gibbs’s 1861, 38, 45–46, 49, 57; Gibbs’s 1863, 38–42, 44, 46, 49, 50, 57, 58, 62–63, 199n38, 199n51; Henry, Gibbs, and Baird’s, 43– 44; Henry’s, 43–44, 46, 51–53, 200n60; Holmes and Mason’s, 53–57, 60, 62–63, 103, 128, 201n105; as immutable mobiles, 60–61; “Indian Languages of North America,” 38, 62, 199n36; legitimize power of early anthropology, 30; Mason on importance of, 95; Mason’s, 47–50, 53, 57, 58, 83, 100, 103, 128; objectivities espoused by, 58–59; Powell’s, 50–51, 199n38; purpose, 29; Rau’s, 53, 201n96; and resistance to collecting, 61–63; Sturtevant’s, 20, 125, 128–32, 133; Thomas’s, 53, 201n96; value of, 33 citizen science. See non-specialist collectors classical episteme, 10, 93, 101, 206n55 classification: approach to, 4–5, 14, 16, 17–19, 20; Baird’s influence on, 37–38, 85; by Bolles, 80–81; in circulars, 34, 42, 44, 49, 50, 54–57, 58, 63; confers value, 97, 101, 110; of culture, 150; Douglas on, 9–10, 45; enduring legacy of, 188; formal vs informal, 158; future research on, 193; of Hoonah Repatriation Collection, 83–84, 203n41, 203n43; by Indigenous peoples, 163, 212n14; of Indigenous peoples, 176; influence of repatriation legislation on, 175; lack of single scheme

of, 124; in ledger books, 63, 71; Linnaean system of, 10, 11–12, 196n34; Mason’s approach to and influence on, 85, 92–93, 95–98, 100–1, 104–5, 123, 206n70; of material types, 144–46; of natural world, 29; in Niblack’s report, 83; as political force and rubric for bureaucracies, 157; pragmatism, digitization, and, 126, 147– 50, 162–63; as purview of experts, 107; role of drawings in, 77; shift away from, 111–12; in situ, 59; in Smithsonian’s annual reports, 97, 205n31, 206n70; Smithsonian’s desire for routinized system of, 92; special interest groups’ work in, 141; Sturtevant’s approach to, 127–33, 148. See also controlled vocabularies; data fields; hierarchies and hierarchical classification; names and naming clerks: Latham, 93, 106, 204n21, 207n77; recording, 103, 104; Riesenberg’s request to hire, 116; type up handwritten cards, 106, 112, 119–20. See also cataloguers Clifford, James, 13 “Coast Indians of Southern Alaska and Northern British Columbia” (Niblack), 83, 203n41 “Code of Symbols for Charts of Prehistoric Archaeology” (Mason), 201n92 coding system, Sturtevant’s, 132–33 collected by (data field), 73, 74(t) collecting. See gaps in collections; material culture collecting; natural history specimens collecting guides. See circulars and field guides collection events (data field), 160 collection name (data field), 160–61 Collections Management and Conserva­ tion, 25 collections managers and staff: collections care shifts to, 134, 137; create data fields, 149; Elder, 164; frustrations with early computerized records, 167; influence of data errors on, 164, 165, 166–67; main tasks today, 25, 169; need for, 156; repatriation legislation and, 177; standardized terms created for, 180; Wilcox, 141, 142–43, 167; work with index terms, 153, 162. See also Greene, Candace

collector (data field), 71, 109–10 collectors: amateurs vs professionals, 202n13; announcements by, 92; as authoritative sources, 50; authority name data field for, 177; credit given to, 48, 52; importance of data from, 70; listed in ledger books, 69; Mason on, 95. See also donors; non-specialist collectors collectors’ lists (field catalogues), 41, 49, 70, 94 Collier, Fred, 140(f ) colonialism: approach to, 4, 5, 7–8, 14–15, 21, 26–27, 185, 188; Duarte and BelardeLewis on, 6; history of anthropology as history of, 13; in Holmes and Mason’s circular, 54; invisible legacy of, 184, 185, 188; of legacy data, 159; in NAGPRA, 193; in offensive terminology, 172 comparisons of specimens, 96, 97 computerized records, early: approach to, 20, 187; catalogue cards underlined for input into, 164, 165(f ); Chenhall on, 210n42; difficulty of inputting card catalogue data into, 150–51, 164, 166–67, 168–69; as means of producing knowledge, 157; and need for inventory of collections, 141–47; Oklahoma Pilot Inventory and, 209n22; pragmatic classification and, 147–50; purpose of inscribing ledger books and card catalogues into, 141, 182; search for errors in, 167–68; Smithsonian recognizes potential of, 134; and updating naming practices, 150–55; value and challenges of, 125–26, 137, 149. See also databases; SELGEM conservation, 144, 145–46 Conservation Lab, 141 conservators, 145 content management systems (CMS), 212n14. See also EMu controlled vocabularies, 148–49, 150, 151– 55, 162, 169. See also names and naming; standards and standardization copies: of ledger books, 72–73; of objects, 22, 190–91, 214n16, 214n18 copyists, 106, 207n77 corresponding number of (data field), 73 cost (data field), 74(t) cradles, 146

Index

231

Croswell, William, 90 culture: classification of, 150; data fields for, 161, 163, 175–76. See also ethnonyms; material culture collecting; tribal affiliation “Culture Terms Not in Use,” 171, 212n31 curators: and amateur ethnologists, 13; authority to classify objects, 92; changing role of, 105, 109, 136–37; current projects in object replication, 214n16; errors and, 111, 135, 164, 165; Gibson, 135, 149–50, 166; Goode, 46–47, 97; indexes kept by, 103, 120, 122; influence of, 61; intellectual nature of position, 156; parameters of work of, 95; Riesenberg, 116–17, 136, 177–78; use of computers, 142; values of, 158; work on card catalogues, 103, 106, 107, 112; work on EMu, 161; work on index terms, 153, 154; work on names, 168, 171. See also Baird, Spencer; Mason, Otis Tufton; Sturtevant, William current number (data field), 73 Dagonet, François, 202n131 Dakota language, 199n36 dances, 22 Darwin, Charles, 196n34 data: anthropological vs conservation, 146; approach to, 4–5, 8, 16, 20, 185, 188; bad or dirty, 20–21, 167–72, 182; catalogue’s “unmediated” access to, 189; circulars and, 42, 44, 51, 53, 58, 63, 70; digitization and, 142, 147, 179–81 (see also data fields); field data, 17–18, 33; generated by researchers, 190; as historical and constructed, 7, 195n14; Indigenous sovereignty of, 199n48; mid-1960s shift in desired, 134–35; non-specialized workers’ responsibility for, 137; standardization of, 141; Sturtevant on importance of, 128–29, 130–31; transfer and migration through time, 163–67; use of term, 7; USNM’s goal of creating large sets of, 9. See also errors and inconsistencies; legacy data; Mason, Otis Tufton: on importance of data; value data coordinators, 141 data entry firms, 137, 164, 168 data fields: in accessioning, 135; in card catalogues, 102, 114, 118(f )–19(f ), 120,

232

Index

121(f ), 124, 150–51, 166; “collector,” 71, 109–10; in databases, 159, 160–61, 162– 63, 167, 169, 175–76, 177, 179–81; in early computerized records, 141, 143–46, 149, 151–55; interdepartmental differences and similarities in, 17, 24, 197n76; in ledger books, 68–75, 78, 94, 118; manifestations of politics in, 186; narrative fields, 191; for object descriptions, 136; “provenance,” 41, 124, 135; used by Repatriation Office, 213n41. See also accession number (data field); catalogue number (data field) data managers, 25, 134, 136, 156, 169, 171 data manuals, 160, 162 databases: approach to, 20–21, 187–88; Greene on limitations of, 176–77, 189– 90; Indigenous-specific, 192; influence of repatriation legislation on, 174–77; names updated in, 168–69; overview of Smithsonian’s, 24; paper media as foundation of, 5; standards of, 159–60, 163, 211n5; thesauri and dictionaries in, 149, 160. See also computerized records, early; data fields: in databases; EMu; INQUIRE Datson, Lorraine, 12–13, 86, 196n42, 196n50 Dauber, Kenneth, 26, 30 Davis, Henry, 45 de-accession. See repatriation deadwork. See routine work decolonization, 6, 186, 191, 194n5 Department of Anthropology (earlier Division of Anthropology): approach to, 4; created, 9, 31, 47; curators and heads, 24, 103, 107, 113, 127, 201n105; differences in divisional records, 112, 113; structure, 24–25, 33, 104, 122, 128, 173; USNM’s decreasing focus on, 105 Department of Arts and Industries, 122 Department of Ethnology (British Museum), 99 Department of Ethnology (NMNH). See Division of Ethnology desiderata: in Baird’s circulars, 35–36; defined, 17, 29; in Gibbs’s circulars, 39, 40–42, 49; in Henry’s circulars, 43–44, 51–52; in Holmes and Mason’s circular,

54–55, 56(t); importance in forming anthropological knowledge, 63; in Mason’s circular, 48–49, 50; in Powell’s circular, 50–51; requested by Davis, 45; types specified in circulars, 33 Dewey, Melvil, 90 dictionaries, 149 difference. See taxonomies Digital Asset Management System (DAMS), 24 digital media. See computerized records, early; databases Digitization Program Office, 22 “Directions for Mound Exploration” (Thomas, 1884), 53, 201n96 directors, 61, 106 dirt, 9–10, 182 dirty data, 20–21, 167–72, 182. See also errors and inconsistencies discursive stability, 15 ditto marks, 69, 78, 82 Division of Anthropology. See Department of Anthropology (earlier Division of Anthropology) Division of Ethnology: conflicting ways of organizing information in, 114; created, 46; curators, 46, 79, 201n105; merger with BAE, 47, 200n76 Division of Physical Anthropology, 104 documentation: approach to, vii, 4–8, 14–15, 16, 26–27, 185; defined in NAGPRA, 3, 5; Hull on, 195n16; legacy and future of, 21, 184–93; methodology of study on, 23–26. See also card catalogues; circulars and field guides; computerized records, early; databases; ledger books donors, 112, 213n41. See also collectors; non-specialist collectors Douglas, Mary, 9–10, 11, 45, 182 Doyle, Ann, 148 drawer (portrait), vi–vii drawings in ledger books, 75–77 dry or alc. (data field), 71 Duarte, Marisa Elena, 6 duplicate objects, 19, 57, 84, 101, 204n46, 206n63 durability. See material durability Duress (Stoler), 184 Dutch museums, 68, 204n47

Elder, Robert, 120, 157, 164 Elders, 116 EMu: accessibility of, 25; comes into use, 177; data fields in, 160–61, 162–63, 169, 177, 179–81; errors transcribed into, 182; online component, 169–72, 179–81, 212n31; precursors, 24, 137, 160. See also databases encyclopedic method, 96–97, 205n47 English language, 38, 42 Entrex, 143, 210n51 epistemologies: approach to, 64, 123, 185; in circulars, 46; classical episteme, 10, 93, 101, 206n55; inscribed and re­ inscribed by documentation, 65–66; Renaissance episteme, 100; shifts in, 107, 112, 127, 155–56. See also natural history errors and inconsistencies —in card catalogues: curators’ frustration with, 135; Hough’s desire to correct, 110; in Key to Collection cards, 120; mid1930s attempts to correct, 111; in naming, 151; Riesenberg on, 116–17; Setzler on, 114–15; in tribal index file, 117, 122; True on, 103; underlining and, 164, 165; use of inventory to correct, 142, 144, 147 —in cataloguing Hoonah Repatriation Collection, 81–82 —as challenge for early computerized records, 125 —drawings as insurance against, 76 —found by Bolles, 80 —made when digitizing card catalogues, 166, 182 —Mason on, 94–95, 205n26 —material durability of, 189 —repatriation and, 159, 174–75, 177 —routine work and potential for, 157 —sources checked for, 167–68 “Eskimo” collections, 96. See also Hoonah Repatriation Collection ethical issues, 181, 188, 191 “Ethnological Directions Relative to the Indian Tribes of the United States” (Mason, 1875). See circulars and field guides: Mason’s ethnologists: cataloguers and curators as amateur, 13; early approach to material culture collecting, 28–29

Index

233

ethnology: approach to, 14, 16, 17; Dauber on, 26; EMu and images of card catalogues from, 180, 181; in Henry’s circulars, 43–44, 46, 200n60; in Holmes and Mason’s circular, 54–57; as science, 33; vs archaeology, 44. See also anthropology; circulars and field guides: Gibbs’s 1863; Division of Ethnology Ethnology File, 141 ethnonyms, 132, 163, 174–78, 180, 209n19 ethnoscience, 128 Europe, 89 European museums: circulars and field guides in, 30; future research on, 192; influence of natural history on anthro­ pological displays at, 11; influence on Mason’s development of card catalogue, 98–99; object exchanges with, 46, 204n47; South Kensington Museum, 205n47; terminology used in, 152, 153; use lists to record objects, 68 European science: anthropology and ethnology as, 8, 33; approach to, 14, 17; Baird’s desire to follow methodology of, 70; Datson and Galison on, 12–13, 86, 196n42, 196n50; evolutionary approach to archaeology in, 196n34; in Holmes and Mason’s circular, 54; Hough on importance of, 110; immutable mobiles spread, 61; importance of drawings in, 77; Indigenous peoples described through language of, 29; Linnaean classification system, 10, 11–12, 196n34; Mason and, 93, 95, 96–97, 101, 105, 205n26; modern classification and, 163; objectivity in, 13, 86; philosophies of, 195n18; Sturtevant on importance of, 129, 131 evidence: approach to, 14, 16, 17; descriptions as, 86; importance of documentation for ensuring, 123–24; objects as, 11, 12; specimens as, 29; types provided by material collections, 110 evolution, 11–12, 29, 196n34 Ewers, John, 165–66 exchanges. See object exchanges expeditions, 28, 31, 43, 53 extinct vs living tribes, 12, 39–40, 44, 45, 48, 59

234

Index

feminist scholarship, 14, 194n3, 197n62 field catalogues (collectors’ lists), 41, 49, 70, 94 field data, 17–18, 33 field guides. See circulars and field guides Field Museum of Natural History, 68–69, 104 field site/station (data field), 75 field specimens. See desiderata; natural history specimens fields. See data fields finding aids, 154 First Peoples. See Indigenous communities and peoples Fitzhugh, William, 165–66 folk classifications, 132 Foreman, Edward, 46–47, 75, 76(f ) Foucault, Michel, 11, 15, 89, 101, 206n55 French language, 46 funding, 144, 192 funerary objects, 3, 22, 172–73, 184 furniture, vii Galison, Peter, 12–13, 86, 196n42, 196n50 gaps in collections, 43, 81, 96, 102–3, 110, 131 Geismar, Haidy, 190 “General Directions for Collecting and Preserving Objects of Natural History” (Baird, 1850), 34–38, 46 geographic names and localities, 80–81, 174–77 geographical-tribal index. See tribal index geological sciences, 202n13 Germany, 98–99 Gessner, Konrad, 89 Gibbs, George, 38, 43–44, 45–46, 49, 57, 199n50. See also circulars and field guides: Gibbs’s 1863 Gibson, Gordon, 135, 149–50, 166 Gitelman, Lisa, 65, 66, 196n51, 197n69 glossaries, 80 Goode, George Brown, 46–47, 97 Grand Ronde, 172 Greene, Candace: on history of accessioning, 202n1; index terms and, 153(f ), 154; on ledger books, 67, 68–70; on limitations of databases, 176–77, 189–90; on

problems with Anthropology Inventory Worksheet, 146 “Guide to Field Collecting of Ethnographic Specimens” (Sturtevant, 1967/1969), 20, 125, 128–32, 133 Gustav Heye Collection, 212n34 Handbook of North American Indians (HNAI), 132, 176, 209n18 handwritten vs typed catalogue cards, 106, 112, 118, 119–20, 122 Harding, Sandra, 194n3 Harvard College Library, 90 Harvard Peabody Museum. See Peabody Museum hats, 22 Hawaiian baskets, 205n26 Hawaiians, 173 Hawley, F.H., 122 Hays-Gilpin, Kelley, 185 headings. See data fields Hennessy, Kate, 195n9, 214n16, 214n18 Henry, Joseph: appointments at Smith­ sonian, 47; circulars by, 43–44, 46, 51–53, 200n60; Davis’s list of desiderata and, 45; on a national museum, 30–31; writes preface to Gibbs’s 1863 circular, 39 Hetherington, Kevin, 11 hierarchies and hierarchical classification, 9–10, 129–30, 148–49, 150, 154–55, 162 Hinsley, Curtis M., 205n44 historians, 195n14 history, 41, 57–58 History of Collection cards, 115, 117, 120, 121(f ) HNAI (Handbook of American Indians), 132, 176, 209n18 Hochman, Brian, 14 holiness, 9–10 Hollinger, R. Eric, 203n43 Holmes, William Henry, 24. See also circulars and field guides: Holmes and Mason’s Hoonah Indian Association, 22–23 Hoonah Repatriation Collection, 18, 22– 23, 78–85, 190, 203n38, 203n43, 204n46 Hooper-Greenhill, Eilean, 101 Hough, Walter, 107, 108(f ), 109–10, 111 how acquired (data field), 74(t)

Hudson’s Bay Company, 46 Hull, Matthew S., 195n16 Human Relation Area Files (HRAF), 132, 149–50, 209n20 human remains: collecting, 35–37, 39–40, 43–44, 52, 62; from collection of Army Medical Museum, 104; physical anthropology in category of, 213n53; preserving, 35, 37, 52, 73. See also Native American Graves Protection and Repatriation Act (NAGPRA) hypertext theory, 77 IBM typewriters, 143, 210n50 immutable mobiles, 60–61 “In Reference to American Archaeology” (Henry, 1878), 51–53 inconsistencies. See errors and inconsistencies index term (data field), 163 index terms: influence of repatriation legislation on, 174, 175, 176–77; limitations of, 180. See also controlled vocabularies indexes and early computerized records, 142–43, 146–47. See also card catalogues Indian File. See tribal index “Indian Languages of North America” (circular, 1852), 38, 62, 199n36 “Indigenous” (as term), 194n4 Indigenous communities and peoples: agency of, 158–59; anthropologists’ fear of harming, 195n11; approach to, 5–7, 16; authenticate museum photos, 116; classification by, 163, 212n14; classification of, 176; collecting human remains of, 35–37, 39–40, 43–44, 52, 62; contemporary artwork from, 212n34; desiderata, circulars, and, 41–42, 48–49, 50–51; difficulties finding objects in collections, 213n36; evolutionary approach to, 196n34; Gibbs supports, 199n50; Gibbs’s three classes of, 39–40; help museums update records, 167; interest in collecting material culture of, 9; “lack” of history of, 57–58; language of European science used to describe, 29; on museums as site of harm and healing, 13; names of creators of objects, 135; naming as

Index

235

oppressive to, 6; naming by, 131–32; new relationship between museums and, 22; offensive and out-of-date terms describing, 169–72; portraits, 108(f ), 109; production of identity of, 186; resistance by, 62–63; search terms used by, 161. See also ethnonyms; extinct vs living tribes; repatriation; tribal index Indigenous data sovereignty, 199n48 Indigenous knowledges, 6, 58, 63, 185, 191, 198n4 Indigenous languages: Chinook Jargon, 172; circulars on, 38, 43, 50–51, 62, 199n36, 199n38; Tlingit, 83, 203n43 Indigenous-specific databases, 192 information. See data Information Retrieval Group of Museums Association, 140–41 information studies, 14, 190, 196n51 INQUIRE: data fields in, 162–63, 175–76; errors transcribed into, 182; as key database, 213n40; precursors and successors, 24, 137, 160. See also databases “Instructions for Archaeological Investi­ gations in the United States” (Gibbs, 1861), 38, 45–46, 49, 57 “Instructions for Research Relative to the Ethnology and Philology of America” (Gibbs, 1863). See circulars and field guides: Gibbs’s 1863 “Instructions to Collectors of Historical and Anthropological Collections” (Holmes and Mason, 1902). See circulars and field guides: Holmes and Mason’s Interior Department, 31 internal record number (IRN), 161 interns, 107, 137, 147, 170, 189 “Introduction to the Study of Indian Languages” (Powell, 1877), 50–51, 199n38 inventories, 68, 141–47, 157, 167–68, 176 inventory code sheets, 144–46, 167 “Inventory of the Collection of Curiosities Bequeathed by Mrs. J.L. van Olden­ barneveld,” 68 inventory teams and specialists, 144, 156 Isaac, Jaimie, 214n16 Japan, 72, 80 Jenkins, David, 11

236

Index

Jesup, Morris, 199n29 Jesup, Thomas Sidney, 37, 199n29 Jewett, Charles Coffin, 90, 204n8 “Kaffir” (as term), 170–71 Kavanagh, Thomas, 153(f ) Kéet S’aaxw (Tlingit Killer Whale Clan Crest hat), 22 Key to Collection cards, 117, 120, 121(f ) Klemm, Gustav, 99, 205n44 Knell, Simon, 202n13 knowledge: approach to, vii, 4, 14, 185; furniture as metaphor for, vii; production of, 101, 105, 110, 157, 186. See also Indigenous knowledges knowledge authority, 8 Krajewski, Markus, 14, 89, 187, 204n8 Krupnik, Igor, 209n18 labelling. See classification; names and naming Lampland, Martha, 158, 159 Landow, George P., 77 languages, 42, 46. See also Indigenous languages Latham, Sarah, 93, 106, 204n21, 207n77 Latour, Bruno, 15, 60–61 Law, John, 15, 182, 197n64 ledger books, 65–86; accessible in Museum Support Center, 25; alternative names for, 204n24; approach to, 18–19, 64, 66, 186–87; card catalogues used with, 69, 102–3, 115, 206n63; checked for errors, 168; data copied into card catalogues, 94; data fields in, 68–75, 78, 94, 118; data omitted from, 63; durability of descriptions from, 164; early computerized records and, 147, 151; early use of, 31, 33; at Field Museum, 68–69; as foundational media technology, 5; future research on, 191–92; Gitelman on, 66; Hoonah Repatriation Collection, 18, 78–85, 203n38, 203n43, 204n46; images in EMu, 160; importance of access to, 190; lack of standardization in, 86; Latham’s work on, 106, 204n21; marginalia in, 74–78; Mason’s concerns about, 148; as means of producing knowledge, 157; origin, 68, 202n4; photos, 72(f ); power of, 85; as primary source of

documentation, 88; purpose of digitization, 182; as trusted medium, 67 legacy data: changes in, 162; as colonial, 159; created, 181; defined, 126; influence on repatriation, 23; inventory raises issue of, 167; use of term, 7, 188; workers’ frustration with, 165 Leiden museums, 68, 204n47 Leipzig, 98–99 Lemov, Rebecca, 14 Lesley, Joseph Peter, 107 librarians, 90, 204n8 libraries, 89–90, 95, 193 Library Bureau, 90 Library of Congress, 14 library science, 90 Linnaean classification system, 10, 11–12, 196n34 “List of Anthropological Objects Trans­ ferred from the Ashmolean to the Pitt Rivers, 1886” (inventory), 68 living tribes. See extinct vs living tribes locality: of archaeological sites excluded from online database, 180–81; data fields for, 71, 72, 160, 162, 166, 213n41; shifts in objects’ physical, 175–76; value of data on, 59–60 Locality-Tribal file. See tribal index Lomatewama, Ramson, 185 lookup lists, 162 Lubbock, John, 196n34 Making Culture Lab, 195n9 Marcus, George, 13 marginalia, 74–78 marine invertebrate ledger books, 71, 72–73 marks (data field), 119 Mason, Otis Tufton: archaeology code system created by, 201n92; articles and monographs published by, 92, 204n16; on Circular 316, 52; colonialism of, 193; concerns about ledger books, 148; credited with adoption of card catalogues, 86, 88–89; on curators’ daily routine, 103; death, 105, 126; desire for cataloguing to become “routine,” 107; on errors, 94–95, 205n26; history and key cards based on models by, 120; Hoonah Repatriation Collection and, 23, 79; on importance of data, 3, 8, 59, 95, 98,

104–5; influence on documentation, 123–24; length of tenure as curator, 8; on names, 99–100, 149; personal secretary of, 106; research practices during time of, 81; on rigorous method of accessioning, 205n29; tribal index file and, 122; on value of context for creating specimens, 203n17; vision of order, 93, 155; vs Sturtevant, 127. See also card catalogues: Mason’s vision of; circulars and field guides: Holmes and Mason’s; circulars and field guides: Mason’s; classification: Mason’s approach to and influence on material culture: approach to, 5, 193; and change in anthropological methods, 126–27, 133; Sturtevant’s approach to, 127–28, 131, 156. See also human remains; objects material culture collecting: Baird’s approach to, 34, 36, 37–38, 69–70; early ethnologists’ approach to, 28–29; Gibbs’s approach to, 40–41; of Hoonah Repatriation Collection, 78–79; interest in Indigenous peoples, 9; Mason and, 95, 123; overview of circulars’ importance for, 18; overview of Smithsonian’s practices, 16–17; shift away from, 111–12; strategic approach to, 13. See also desiderata; human remains: collecting material durability: approach to, 14–15, 21, 126, 185; of descriptions, 164; manifestations of politics in, 186; of names, 188– 89; produces dirty data, 182; repatriation and, 179; tension with flexibility, 89, 116, 162; tension with performativity, 99, 124; use of term, 16 material type (data field), 69, 141, 144–46, 163 McCoy, George D., 109(f ) McIntyre, Tom (Mac), 139(f ) media history, 14, 66, 196n51 Merrill, William, 127 Mexico, 80 Meyer, A.B., 90 Microsoft Access, 213n40 misinterpretation. See errors and inconsistencies missing links. See gaps in collections Mol, Annemarie, 15–16 monographic catalogues, 205n47

Index

237

Mooney, James, 115 Moser, Stephanie, 76–77 Mukurtu CMS, 212n14 Munns, Denise, 139(f ) Murdock, George, 209n20 Murphy, Mrs. N.B., 114 museum aides. See assistants and aides Museum Association, 104 Museum Cataloging in a Computer Age (Chenhall), 210n42 “Museum Catalogue Book,” 71 Museum Computer Network, 140 Museum Handbook (National Parks Service), 132, 209n20 museum number (data field), 73, 94, 114 Museum of Ethnology, 98–99 Museum of the American Indian, 212n34. See also National Museum of the American Indian (NMAI) Museum Period, 13 Museum Support Center (MSC), 25, 87, 167 Museum Volkenkunde, 68, 204n47 museums: ethical issues in contemporary, 181; literature on non-neutrality of, 13; professionalization of discipline of, 104. See also American museums; documentation; European museums; Smith­ sonian Institution Museums Journal, 104 musical instruments, 83, 100–1, 203n41 Musical Instruments File, 88, 120, 122 NAGPRA. See Native American Graves Protection and Repatriation Act (NAGPRA) names and naming: approach to, 16, 19, 20; Bolles’s conventions of, 80; of cataloguers, 161; as challenge for early computerized records, 125–26; as continuing instrument of colonialism, 6; controlled vocabularies for, 148–49; curators review, 168; of data fields, 75; Douglas on, 9–10; durability of, 188–89; ethnological collections elude simple standards of, 163; expansion in number of categories for, 167; historical and updated practices of, 150–55; in Hoonah Repatriation Collection, 81–82; by Indigenous communities, 131–32; of Indigenous creators

238

Index

of objects, 135; influence of repatriation legislation on, 174–77; interdepartmental differences in, 160; lack of single scheme of, 124; manifestations of politics in, 186; Mason on, 99–100, 149; meant to help researchers and in-house staff, 180; narrative fields and, 191; offensive and out-of-date, 169–72, 179, 189, 212n31; requested in Gibbs’s circulars, 41; Riesenberg on, 178; role in objectivity, 59; as world building, 158. See also classification; ethnonyms; object name (data field); standards and standardization narrative fields, 191 National Anthropological Archives, 25, 107 National Institute for Promotion of Science, 31 National Museum of Natural History (NMNH): as active research site, 192; approach to, 4, 16; buildings, 141; created, 33; departments and units in, 24, 138, 173; historical ethnological collections in, 212n34; name change to, 194n1. See also United States National Museum (USNM) National Museum of the American Indian (NMAI), 172, 192, 212n34 National Museum of the American Indian Act (NMAIA): approach to, 21; categories of objects in, 172–73, 184; influence on naming, 174–77; influence on standardization, 159; process of repatriation under, 178–79; repatriation of Tlingit hat under, 22; requirements of, 158, 172– 73, 213n36 “National Museum Register – Marine Invertebrates,” 71, 72 National Parks Service (NPS), 132, 209n20 Native American Graves Protection and Repatriation Act (NAGPRA): approach to, 21; and assumptions about museums’ authority, 8, 195n19; categories of objects in, 172–73, 184; colonialism of, 193; definition of documentation in, 3, 5; influence on naming, 174–77; influence on standardization, 159; process of repatriation under, 178–79; requirements of, 158, 172–73, 213n36

Native American peoples. See Indigenous communities and peoples natural history —classification scheme in, 75 —and development of SELGEM, 138 —influence: on anthropology generally, 10, 11–13, 16; on early anthropologists’ critiquing of objects, 58; on early circulars, 46; on early ethnologists’ collecting practices, 29; on ethnology, 95, 205n44; on Mason, 101; on Smithsonian’s collecting practices, 18, 34, 49 natural history specimens: Baird’s approach to collecting, 34–38, 46, 69–70; drawings of, 76–77; in Gibbs’s circulars, 42; in Holmes and Mason’s circular, 54; ledger books of, 71, 72–73; “type specimens,” 11. See also desiderata nature of object (data field), 73 negative number (data field), 119(f ) Niblack, Albert Parker, 83, 203n41 Nichols, Catherine, 30–31, 49, 84, 101 NMAI (National Museum of the American Indian), 172, 192, 212n34 NMAIA. See National Museum of the American Indian Act (NMAIA) NMNH. See National Museum of Natural History (NMNH) NMNH Information Technology Office (ITO), 24 nomenclature. See names and naming non-specialist collectors, 33–34, 39, 45, 52–53, 58, 202n13. See also Bolles, Timothy Dixon; collectors; donors non-specialized workers: Greene on, 189; hired in 1960s, 137; Riesenberg’s request to hire, 116; type up handwritten cards, 119–20; work on card catalogues, 105–7, 111; work on early computerized records, 143–44, 147, 156, 164 North American collections, 148, 152 notes (data field), 181 “Notes and Queries” pamphlets, 30, 200n71 number of specimens (data field), 71, 119, 121(f ) numerical card catalogue, 114

museums, 46, 204n47; of items in Hoonah Repatriation Collection, 84; Mason’s role in, 101; mentioned in Smithsonian’s annual reports, 206n63; role of documentation in, 57; with schools, 107 object name (data field): in card catalogues, 150–51, 155, 163–64, 166; in early computerized records, 141; lack of use by Repatriation Office, 213n41; standardization of, 151–55 “Object of Research in Museums,” 130 object replication, 22, 190–91, 214n16, 214n18 object type (data field), 69 objectivities: approach to, 16, 17; data’s lack of, 7; Datson and Galison on, 12–13, 86, 196n50; espoused by circulars, 58–59; ocularcentric approach to, 198n4; origins, 5; Sturtevant and, 125, 131, 156 objects: as evidence, 11, 12; of knowledge, 66, 67, 81, 124; museum catalogues as, 186; replace human objects, 13; types in repatriation legislation, 3, 22, 172–73, 184; unlocatable, 142, 168, 189. See also desiderata; material culture; rattles; specimens O’Brien, Meghann, 214n16, 214n18 ocularcentrism, 12–13, 29, 61, 77, 198n4, 202n131 offensive terms, 169–72, 179, 189, 212n31 Office of the Chief Information Officer (OCIO), 24 Office of the Registrar, 92 Ogle, Mrs., 111 Oklahoma Pilot Inventory, 132, 136, 209n22 online database, 169–72, 179–81, 212n31 order: Mason’s vision of, 93, 155; as model of rationality, 11; Western assumptions of, 10 original number (data field), 71, 73 ornithology, 69, 202n8 “Outline of World Cultures” (Murdock), 209n20 out-of-date terms, 169, 170–71, 179, 189 Oxford museums. See Pitt Rivers Museum

object descriptions, 134–36, 177–78 object exchanges: card catalogue’s value for, 102; defined, 19; with European

Paper Cadavers (Weld), 196n51 Paper Knowledge (Gitelman), 65, 196n51, 197n69

Index

239

paper media. See card catalogues; circulars and field guides; ledger books Parezo, Nancy, 34, 59 Parry, Ross, 137–38, 140 Patent Office, US Government, 31 patriarchy, 12, 196n36 Peabody Museum, 13, 68, 70, 204n47 performance and performativity: approach to, 15–16, 21; contemporary databases as, 182; of documentation media, 165; manifestations of politics in, 186; as repetition of colonial encounter, 159; scholarship on, 197n62, 197n64, 197n66; tension with durability, 99, 124; by Tlingit dancers, 21–22 Perry, Matthew, 72, 202n16 Petroff, Ivan, 80 Philippines, 53–54, 60 philology. See circulars and field guides: Gibbs’s 1863 physical anthropology, 104, 180, 213n53 Pilot Study for Inventorying Ethnological Collections, 136 Pitt Rivers Museum, 11, 68, 98–99, 152, 153 Plains Indian cradles, 146 Poovey, Mary, 195n13 postcolonialism, 14, 185 Povinelli, Elizabeth, 185 Powell, Wesley, 50–51, 199n38 power: of bureaucracy, 30; future research on, 186; and Indigenous peoples’ resistance, 62–63; of ledger books, 85; of museums in repatriation, 8, 179, 183, 184, 195n19; of naming, 6; postcolonial questions about, 185 Powys, Marian, 78 pragmatism: in accessioning, 134; approach to, 20, 21; of computerization, 141; of data fields, 143–44, 175; digitization, classification, and, 126, 147–50, 162–63; of index terms, 154–55; and offensive terms, 172 Pratt, Mary Louise, 10, 11–12 “Preferred Index Specimen Names,” 152, 153(f ) “Preliminary Catalogue of the Eskimo Collection” (Bolles), 79–81 preparation rooms, 67(f ), 68, 92 prepared by (data field), 73

240

Index

preservation: of specimens, 34–35, 37, 52, 71, 73, 107; writing and, 65 Processing Lab, 141, 157, 168 professionalization, 104 “Projected Computer Programming Sup­ port for Collections Management,” 168 provenance: as data field, 41, 124, 135; effect of inconsistencies on, 165–66; Mason’s circular and, 50; repatriation legislation and, 174; source of data on, 151 punch cards, 14. See also SELGEM Purity and Danger (Douglas), 9–10 Putnam, Frederick Ward, 13, 70 Q?rious gallery, 22 racist terms. See offensive terms rattles: catalogue card for, 165(f ); classified as musical instruments, 83, 100–1, 203n41; conservation data about, 146; data fields for different names of, 152; in Hoonah Repatriation Collection, 80–81, 82–84, 203n43; lookup list for, 162; repatriation legislation and classification of, 175; Riesenberg on, 177–78 Rau, Charles, 47, 53, 201n96 Raven’s Tail-Chilkat, 214n16 Reardon, Jenny, 39 received from (data field), 71, 73, 74(t) “Recent Advances in Museum Method” (Goode), 97 Reciprocal Research Network, 214n20 “Recommendations for Catalog Descrip­ tions,” 134 reconciliation, 168, 192 recording books. See ledger books records. See documentation Recovering Voices program, 192 registrars, 94, 137, 162 remarks (data field), 78 Renaissance episteme, 100 repatriation: colonialism in process of, 8; complexity of, 4; governed by categorical distinctions, 182–83; importance of cultural affiliation in cases of, 161; influence of data legacies on, 66; as part of Smithsonian’s daily work, 192; role of object replication in, 22, 191, 214n18; showcases failure of documentation system, 188; standardization and, 150;

successful cases, 195n17. See also Hoonah Repatriation Collection; National Museum of the American Indian Act (NMAIA); Native American Graves Protection and Repatriation Act (NAGPRA) repatriation lists, 174 Repatriation Office, 22–23, 173, 179, 213n41 replication, object, 22, 190–91, 214n16, 214n18 Report on the Cheyenne-Arapaho Inventory Project (Elder), 157 Research and Collection Information System, 24 researchers: card catalogues designed to benefit, 96, 100, 101–2, 105, 109, 124; collections management shifts to, 134; data fields benefit, 163; data generated by, 190; early computerized records designed to benefit, 149, 154, 164, 168; EMu designed to benefit, 181; influence of data errors on, 167; knowledge of, 191; ledger books designed to benefit, 85; standardized terms created for, 180; training for, 192; work on card catalogues, 112; work on classification, 81, 95; work on EMu, 161 resistance, 61–63 Riesenberg, Saul, 116–17, 136, 177–78 Riles, Annelise, 8 robes, 214n16 Rosenberg, Daniel, 7, 195n13 routine work: in 1960s, 134, 136–37; approach to, 16, 19–20; on card catalogues, 105–11, 120; current belief that cataloguing is, 189; on early computerized records, 144, 167; embedded into information infrastructure, 181–82; influence on repatriation, 179; on inventorying, 157; on ledger books, 70; normalization of cataloguing practices through, 186 Russia, 43 sacred objects, 172–73, 184 “SAI-wash” (as term), 171–72 Savage Kin (Bruchac), 7, 184 science. See European science scientific value. See value Second World War, 114 secretaries, 70, 106

SELGEM: and computerized inventory, 142, 143; development of, 137–41, 156; errors transcribed into, 182; generates list of index terms, 152; and production of knowledge, 157; spelling variations as issue for, 166; successors, 137, 160. See also computerized records, early Sellers, F.H., 69 Setzler, Frank M., 113(f ), 114–15, 116–17 sex (data field), 73, 78 shaman, 78 similitude. See duplicate objects; taxonomies “Sioux” (as term), 176–77 “Siwish” (as term), 171–72 Sly, Conrad, 214n16, 214n18 Smithson, James, 30 Smithsonian Directive 609, 213n53 Smithsonian Institution: as active research site, 192; created, 16, 30; important time of change for, 89; mandate, 30; structure, 24. See also annual reports, Smithsonian’s; National Museum of Natural History (NMNH) Smithsonian Institution Archives, 25 Smithsonian Institution Exhibits, 22, 24 Smithsonian Miscellaneous Collections, 33, 34, 43 Smithsonian Office of Anthropology (SOA), 128 sociologists, 15, 195n14 somatological collections, 54 South Africa, 170–71 South Kensington Museum, 205n47 Spanish language, 38, 42 specimens: comparison of, 96, 97; data input into early computerized records from, 143–44, 166; as evidence, 29; making of, 155; objects become, 18, 66, 81, 86, 97, 123–24, 129, 130, 186. See also desiderata; natural history specimens; objects; preservation: of specimens; value speculative collecting, 59–60 Spinner, John, 139(f ) “squaw” (as term), 170 staff. See workers standards and standardization: approach to, 20; and clean data, 169; for databases, 159–60, 163, 211n5; digitization,

Index

241

pragmatism, and, 147–50; formal vs informal, 158; future research on, 193; imposed by material forms, 146; influence of NMAIA and NAGPRA on, 159; ledger books’ lack of, 86; limitations of, 180; necessary with computerized records, 137–38, 140–41; offered by card catalogues, 91. See also controlled vocabularies; names and naming standpoint theory, 194n3 Star, Susan Leigh, 14, 75, 157–58, 159 station (data field), 73, 75 stenographers, 106 Stoler, Ann, 15, 75, 184, 185 storage: of card catalogues, 114; collections moved to new facility, 122–23, 141; current management of, 25; use of collections inventory to improve, 141, 143, 144, 145, 147 students, 144 “studying up” (as term), 194n3 Sturtevant, William: appointed curator, 127; approach to classification, 127–33, 148; on artifacts as concrete objects of culture, 155; circular by, 20, 125, 128–32, 133; influence of, 156; work on HNAI, 132 “Suggestions Relative to Objects of Scientific Investigation” (Henry, Gibbs, and Baird, 1867), 43–44 Suitland storage facility. See storage Summer School in Museum Anthropology, 192 Swan, James, 52 synonyms, 151, 152, 153 “Systema Natura” (Linnaeus), 10 TallBear, Kim, 39 taxonomies: of similitude, 82, 84, 97–98, 100–1; of similitude vs difference, 11, 81. See also standards and standardization terminologies. See names and naming “Terms and Conditions for Use of Online Collections Databases,” 213n53 thesauri, 149, 151, 152, 153(f ), 160 Thomas, Cyrus, 53, 201n96 Thompson, Jann, 140(f ) 3D models, 22, 190–91, 214n16, 214n18 Tlingit Killer Whale Clan Crest hat, 22 Tlingit language, 83, 203n43

242

Index

Tlingit peoples, 21–22, 176. See also Hoonah Repatriation Collection tribal affiliation, 34, 55, 75, 159, 175–76, 198n20. See also culture tribal index: errors in, 117, 122; photo, 88(f ); purpose, 87, 88, 120; reflects Mason’s vision of card catalogues, 112; revision of, 113–14, 117, 122 tribal names. See ethnonyms tribal-land geographical index. See tribal index Tribal-Locality File. See tribal index tribe (data field), 72, 75 True, J.W., 103 typed catalogue cards. See handwritten vs typed catalogue cards typists, 106, 112, 116, 119–20, 141, 167 typological tradition, 11 union catalogues, 131, 136 United Kingdom and Britain, 30, 104, 141, 200n71 United States: Museum Period in, 13; professionalization of museum discipline in, 104. See also American museums, other; National Museum of the American Indian Act (NMAIA); Native American Graves Protection and Repatriation Act (NAGPRA) United States National Museum (USNM): BAE’s material transferred to, 200n76; buildings, 31–33, 47, 89, 105; created, 46, 89; goal of, 9; heads, 46, 198n22; name change from, 194n1; storeroom in, 91(f ); subdivided, 33. See also National Museum of Natural History (NMNH) University of Oklahoma, 209n22 unlocatable objects, 142, 168, 189 Urban, Richard, 106 Urry, James, 33 US Congress, 30–31, 144, 172 US Exploring Expedition, 31 value: approach to, 185, 188; card catalogues confer, 102; classification confers, 97, 101, 110; of collections tied to value of data, 142–43; of context for creating specimens, 203n17; of data for creating specimens, 3, 59–60, 116, 123–24, 130; documentation confers, 57, 67, 124; of

Elders’ information, 116; museums’ responses to systems of, 158; of similar vs unique objects, 97. See also desiderata vandalism, 59–60 visual media, 14 visual pedagogies. See ocularcentrism volunteers, 189 Washington, DC, 30, 140 Waterton, Clare, 16, 197n66 Weld, Kirsten, 196n51 “Whatsit” file, 152 when collected (data field), 73, 74(t) when entered (data field), 74(t) Wilcox, Vincent, 141, 142–43, 167 Wilkes collection, 31 Wilson, Jordan, 195n9 women workers: Bender, 120; Latham, 93, 106, 204n21, 207n77; Murphy, 114; Ogle, 111; Powys, 78. See also Greene, Candace work roles, changing, 105, 109, 133–34, 136–37, 156 workers: administrators, 92; computer­ ized records designed to benefit, 149;

conservators, 145; and limitations of data amounts, 147; on limitations of fields in databases, 159, 166; marginalia by, 74–78; on offensive terminology, 170–71; photos, 67(f ), 108(f ), 139(f )– 40(f ); shortages, 116, 119–20, 167; work logging objects in books, 67(f ), 68. See also assistants and aides; cataloguers; clerks; collections managers and staff; curators; non-specialized workers Works Progress Administration (WPA), 106 “Wrapped in the Cloud” (collaborative project), 214n18 Wright, Alex, 14, 87 writing and preservation, 65 written history, 41 Wumpurrarini-kari Archive, 212n14 wunderkammern (cabinets of curiosity), vii, 11 Yale University, 209n20 zoological ledger books, 71, 72–73

Index

243